Cloud Hosting

Image generated by Author with the help of DALL-E

I spent the past few months trying to build all sorts of LLM-powered apps, and truthfully, a really significant portion of time was just dedicated to improving prompts to get my desired output from the LLM.

There have been many moments where I run into a sort of existential void, asking myself if I might just be a glorified prompt engineer. Given the current state of interacting with LLMs, Im still inclined to conclude with Not Yet, and on most nights, I overcome my imposter syndrome. Wont get into that today.

But I still often wonder if, one day, the process of writing prompts could be mostly automated away. And I think the answer to this futuristic scenario hinges on uncovering the nature of prompt engineering.

Despite the countless number of prompt engineering playbooks out there on the vast internet, I still cannot decide if prompt engineering is an art or a science.

On one hand, it feels like an art when I have to iteratively learn and edit my prompts based on what Im observing in the outputs. Over time, you learn that some of the tiny details matter using must instead of should, or placing the guidelines towards the end instead of the middle of the prompt. Depending on the task, there are simply too many ways that one can express a set of instructions and guidelines, and sometimes it feels like trial and error.

On the other hand, one could argue that prompts are just hyper-parameters. At the end of it, the LLM really just sees your prompts as embeddings, and like all hyper-parameters, you can tune it and objectively measure its performance if you have an established set of training and testing data. I recently came across this post by Moritz Laurer, whos an ML Engineer at HuggingFace:

Every time you test a different prompt on your data, you become less sure if the LLM actually generalizes to unseen data Using a separate validation split to tune the main hyperparameter of LLMs (the prompt) is just as important as train-val-test splitting for fine-tuning. The only difference is that you dont have a training

See the rest here:

Automated Prompt Engineering. A mixture of reflections, lit reviews | by Ian Ho | Mar, 2024 - Towards Data Science

Read More..

Cell culture

Human embryonic kidney (HEK)293E (ATCC, CRL-1573), Raji cells (ATCC, CCL-86), WSU-NHL (DSMZ GmbH, ACC 58), DOHH-2 (DSMZ GmbH, ACC 47) and SU-DHL-4 (DSMZ GmbH, ACC 495) were maintained in RPMI 1640 medium with L-glutamine (ThermoFisher) all supplemented with 10% heat-inactivated (HI) fetal bovine serum (FCS) (Merck), 25g/mL streptomycin and 25U/mL penicillin (ThermoFisher). T84 cells (ATCC, CCL-248) were maintained in HAMs F12/DMEM (1:1) with L-glutamine (ThermoFisher) supplemented with 10% HI-FCS, 25 g/mL streptomycin and 25U/mL penicillin (ThermoFisher). HMEC1-hFcRn cells63 were maintained in MCDB 131 medium (ThermoFisher) supplemented with 10% HI-FCS, 2 mM L-glutamine, 25 g/mL streptomycin, 25U/mL penicillin, 10ng/mL mouse epidermal growth factor (mEGFR) (Peprotech) and 1g/mL hydrocortisone. MDCK-hFcRn cells were generated Drs. Jens Fisher and Alex Haas (Roche Pharma Research and Early Development) and cultured in DMEM supplemented with 10% HI-FCS, 25 g/mL streptomycin, 25U/mL penicillin, and 300g/mL G418 (Sigma-Aldrich). Expi293 cells (ThermoFisher, A14527) were maintained in Expi293 medium (ThermoFisher) supplemented with 10% HI-FCS, 25g/mL streptomycin and 25U/mL penicillin (ThermoFisher). All cell lines were kept in a humidified 37C/5% CO2 incubator, except Expi293 cells that were kept in a humidified 37C/8% CO2 incubator on an orbital shaker (125rpm).

Expression vectors encoding the heavy chain (HC) and light chain (LC) of NIP64, anti-CD20 (mAb2, mAb1, mAb9 and ofatumumab)42,65, anti-S. aureus WTA (4497)46, anti-S. pneumonia (Dob1)48 and anti-N. gonorrhoeae (2C7)66 specific mouse-human chimeric or human WT IgG variants were generated by synthesizing cDNA followed by subcloning into the described pLNOH2/pLNOk vector system (Genscript Inc). SARS-CoV-2 specific mAb4 (THSC20.HVTR04)33 WT IgG1 were generated by synthesizing the V-region followed by subcloning into the pFUSE expression vector system (Invivogen) (Genscript Inc). Vectors encoding IgG variants with site-specific substitutions were generated by site-directed mutagenesis (Genscript Inc). All IgG variants or recombinant Fc fragments were produced in HEK293E cells or Expi-CHO (Humab 2C7) by transient transfection using Lipofectamine 2000 (ThermoFisher). The IgG variants were purified using CaptureSelect CH1 columns (Thermo Fisher) prior to size exclusion chromatography (SEC) to isolate monomeric fractions using a Superdex 200 Increase 10/300 column (Cytiva Life Sciences) with an KTA Avant 25 (Cytiva Life Sciences). Purified IgG variants were concentrated using Amicon Ultra 50K spin columns (Merck) and stored in either 1x phosphate buffered saline (PBS) (NIP, anti-CD20, anti-S. aureus WTA (4497), anti-S. pneumoniae 6B (CPS6) and anti-N. gonorrhoeae (2C7) or 20mM TRIS-HCl, 140mM NaCl, pH 5.6 (mAb4). Protein concentrations were determined using a DS-11 spectrophotometer (DeNovix). For production yield experiments, NIP IgG1 WT and REW were produced in Expi293 cells using the Expifectamine Transfection Kit (Thermo Fisher) and purified as above.

Truncated soluble and Glutathione S-transferase (GST) tagged human FcRn lacking the transmembrane domain (hFcRn-GST) was produced in HEK293E cells and purified on a GSTrap column (Cytiva Life Sciences)67. Truncated soluble and biotinylated human and mouse FcRn formslacking the transmembrane domain were acquired from Immunitrack Inc. Truncated soluble His6x tagged human FcRn lacking the transmembrane domain (hFcRn-His) was produced in a Baculovirus expression system and purified using a HisTrap HP column (Cytiva Life Sciences)68,69. The Baculovirus stock was a kind gift from Dr. Sally Ward (University of Southampton). Monomeric fractions of hFcRn-His were isolated by SEC using a Superdex 200 Increase 10/300 column (Cytiva Life Sciences) with an KTA Avant 25 (Cytiva Life Sciences).

Stated titration series of NIP specific IgG variants were captured on bovine serum albumin (BSA) conjugated to NIP (1:25 ratio) (BSA-NIP(25)) coated at 1g/mL/100l in 96-well EIA/RIA plates (Corning) and blocked with 4% skimmed milk (S) in PBS overnight (ON). All remaining steps were performed using either phosphate buffer pH 6.0 with 4% S and 0.05% Tween 20 (T) or PBS/T/S pH 7.4 as dilution and wash buffers. Then FcRn-GST (0.25g/ml or 2.0g/mL), biotinylated human or mouse FcRn (Immunitrack) (2.5g/mL or 0.25g/mL) were added and incubated for 1h at RT. When 0.25g/mL biotinylated FcRn were used at pH 7.4, the receptor was pre-incubated with alkaline phosphatase (AP) conjugated streptavidin (1:5000 in PBS/T/S) to increase sensitivity. Bound FcRn was detected either by a horseradish peroxidase (HRP)-conjugated anti-GST antibody from goat (Rockland Immunochemicals, 200-301-200) (diluted 1:5000 in PBS/T/S) or streptavidin-AP (diluted 1:5000 in PBS/T/S) and visualized by addition of TMB substrate (CalBiochem) or p-nitrophenylphosphate (AP substrate) (Sigma-Aldrich) diluted to 10g/mL in diethanolamine buffer. The HRP reaction was terminated by addition of 1M HCl and the 450nM (HRP) or 405nm (AP) absorption values was recorded using a Sunrise Spectrophotometer (TECAN).

A Biacore 3000 (Cytiva Life Sciences) was used to couple NIP IgG1 variants (300 resonance units (RU) to CM5 sensor chips using an amine coupling kit (Cytiva Life Sciences). Phosphate pH 6.0 or HBS-P+ pH 7.4 were used as running and regeneration buffers, respectively. Serial dilutions of monomeric hFcRn-His were injected over immobilized mAbs at pH 6.0 with a flow rate of 50l/min at 25C. Binding data were adjusted to a zero sample and the reference flow cell values subtracted before the Langmuir 1:1 ligand binding model (BIAevaluation software) were used to determine binding kinetics. Binding at pH 7.4 was performed by single injections of 4000nM FcRn-His over 2000 RU of immobilized NIP IgG1 variants at 25C with a flow rate of 20l/min.

HMEC-1 cells stably expressing HA-hFcRn-EGFP were seeded into 24-well plates (CorningCostar) at 7.5105 cells/well and cultured for 2 days in complete growth medium. The cells were washed twice and starved for 1h in Hanks Balanced Salt Solution (HBSS) (ThermoFisher). Then, 200nM NIP specific IgG1 variants diluted in 250l HBSS (pH 7.4) were added in triplicates to two identical plates of cells and incubated for 4h in a 37C/5% CO2 incubator. The HBSS was removed, and the cells washed four times with ice cold HBSS (pH 7.4), before fresh growth medium without FCS and supplemented with MEM non-essential amino acids (ThermoFisher) was added to one of the plates and incubated ON before samples were collected (recycling fraction). Total protein lysate (residual fraction) was then isolated using RIPA lysis buffer (ThermoFisher) supplied with 1x Complete Protease Inhibitor (Roche). The mixture was incubated with the cells on ice and a shaker for 10min followed by centrifugation for 15min at 10000g to remove cellular debris. Similarly, total protein lysate (uptake fraction) from cells in the second plate was isolated after 4h using the same protocol. The amounts of NIP IgG1 variants present in the samples and lysates were quantified by ELISA.

Plasma half-life experiments were performed by Jackson Laboratory Services (Bar Harbor, USA). Hemizygous FcRn transgenic mice (B6.Cg-Fcgrttm1Dcr Tg(FCGRT)32Dcr/DcrJ) (Tg32) that are knockout for the mouse FcRn heavy chain (Fccgrttm1Drc) and express the genomic transgene of the human FcRn heavy chain (FCGRT) under the control of the human FcRn promotor (Tg32) (The Jackson Laboratory) were used to measure the plasma half-life of IgG variants and etanercept. A mix of 3 female and 2 male mice aged 79 weeks was used per group. The mice were pre-loaded with 500mg/kg IVIg (privigen, CLS Behring) via i.v. administration 2 days prior to i.v. administration of the test antibodies at a dose of 5mg/kg. Blood samples (25 L) were drawn from the retro-orbital sinus at days 1, 2, 3, 5, 7, 10, 12, 16, 19 and 23 days post administration of the test antibodies. Blood samples were mixed with 1L 1% K3-EDTA to prevent coagulation and centrifuged at 17.000g for 5min at 4C. Plasma was isolated and diluted 1:10 in 50% glycerol/PBS solution before stored at 20C until analysis by ELISA. Half-life data were plotted as percent antibody remaining compared to the first concentration measured. Data points from the -phase were used to calculate half-life using the formula:

$$,tfrac{1}{2}=frac{log left(0.5right)}{log left(frac{{Ac}}{{Ao}}right)}x,t$$

(1)

where t1/2 is the half-life of the antibody, Ac is the amount of antibody remaining, and A0 is the original amount of antibody at day 1 and t is the elapsed time70. Where stated NCA PK model parameters were determined from the measured antibody concentrations in plasma using gPKPDsim for MatLab34. Following the final plasma sample collection on day 23, the mice were overdosed with tribromoethanol. Gauge needles were inserted into the trachea of each mouse to slowly inject 1mL PBS into the lungs before withdrawn 2 times. BALF samples were stored at 20C until analysis. Half-life experiments in WT Balb/c mice (Taconic Farms) (6 mice/group, 8 weeks old) were performed at Oslo University Hospital animal facility. NIP IgG1 WT and REW were administered i.v. at 5mg/kg and blood samples collected by puncture of the saphenous vein and collected using heparinized micro capillary pipettes at day 1, 2, 3, 4, 7, 8, 11, and 15 post injection. Half-life was calculated as above. The experiments were approved by the Norwegian Food Safety Authority.

96-well EIA/RIA plates were coated with anti-NIP IgG1 WT, anti-NIP IgG1 REW, anti-SARS-CoV-2 IgG1 WT and anti-SARS-CoV2 REW (mAb4) (diluted to 1g/mL/100l per well). In addition, mouse IgG from serum (Sigma-Aldrich, I5381) was coated in parallel (0.25g/mL) as a positive control for the detection antibody used. Then, plasma samples collected at all timepoints of the plasma half-life experiments, including pre-bleed samples, were diluted 1:200 and added to the plates followed by incubation for 1h at RT. Captured mouse IgG was detected using an AP-conjugate anti-mouse IgG (Fc-specific from goat (Sigma-Aldrich, A9316) (diluted 1:5000 in PBS/T/S) and visualized by addition of AP-substrate (10 ug/mL in diethanolamine buffer) (Sigma-Aldrich). Absorbance values were recorded at 405nm using a Sunrise Spectrophotometer (TECAN).

Homozygous B6.Cg-Fcgrttm1Dcr Tg(FcGRT)32Dcr/DcrJ mice (Tg32) (The Jackson Laboratory) (6 mice/group, 8 weeks old) were used. A mix of 3 female and 3 male mice was used per group. When sedated after intraperitoneal delivery of ZRF cocktail (250mg/mL of Zoletil Forte, 20mg/mL of Rompun, 50g/mL of Fentanyl), 10l of NIP IgG1 WT or REW diluted in PBS (2.23mg/kg) was given to each nostril followed by inbreath while lying on their backs. Blood was collected by puncture of the saphenous vein and collected using heparinized micro capillary pipettes 24h post administration and analyzed by ELISA. The experiment was performed at Oslo University Hospital animal facility and approved by the Norwegian Food Safety Authority.

Assayswere performed using Transwell filters (1.12cm2) with collagen coated polytetrafluoroethylene (PTFE) membranes and 0.4m pore size (Corning Costar). The filters were incubated ON in complete growth medium followed by seeding of 1.0106 T84 or 1.4106 MDCK-hFcRn cells per well. Transepithelial resistance (TEER) was monitored daily using a MILLICELL-ERS-2 volt-ohm meter (Millipore). The T84 cultures were grown for 45 days before reaching confluency with a TEER value of 10001200 x cm2 while MDCK-hFcRn were grown for 24h before reaching a TEER value of 600800 x cm2. Prior to experiments, the cells were starved for 1h in HBSS. Then 200nM (200l) of NIP-specific IgG1 variants were added to either the apical or basolateral reservoir and incubated at 37C for 4h before samples were taken from the opposite reservoir and analyzed by ELISA.

The cDNA fragment encoding amino acids 18541 of hemagglutinin (HA) from influenza A H1N1 (A/Puerto Rico/8/1934 (PR8)) was used to generateIgG1 Fc fragments with HA fused to the N-terminal end of one of the HCs. The HA cDNA was subcloned into the pFUSE-hIgG1-Fc2 expression vector (InvivoGen) to generate pFUSE-HA(PR8)-hIgG1-Fc2. Removal of the multiple cloning sites in the target vector generated pFUSE-naked-hIgG1-Fc2. Mutagenesis was then performed to introduce the knob-in-hole mutations. The knob mutation (T366Y) was introduced into pFUSE-HA(PR8)-hIgG1-Fc2, while the hole mutation (Y407T) was introduced into the pFUSE-naked-hIgG1-Fc construct either alone or in combination with the REW substitutions. The monovalent Fc fusions were produced in Expi293 cells by transient transfection adding a 2:1 ratio of the knob:hole constructs per the manufacturer instructions. The fusions were purified using a CaptureSelect FcXL affinity matrix (ThermoFisher) packed in a 5ml column (Repligen) per the manufacturer recommendations. Eluted fractions were collected, concentrated and buffer exchanged to 1xPBS using Amicon Ultra-30 spin columns (Merck). Monomeric fractions of the monovalent fusions were then isolated by SEC using a Superdex 200 Increase 10/300GL column (Cytiva Life Sciences) with an KTA Avant 25 instrument (Cytiva Life Sciences).

96-well EIA/RIA plates (Corning) were coated with a recombinant human albumin variant (8g/mL in PBS/ 100L per well) engineered to bind pH independently to human FcRn29. The plates were blocked with PBS/S and washed before recombinant soluble human FcRn-His was added and incubated at RT for 1h. Then, stated titrated amounts of monovalent HA Fc-fusions or anti-NIP IgG1 variants were diluted in PBS/T/S pH 6.0 or pH 7.4 and added to the plates. All remaining steps were performed using PBS/T/S pH 6.0 or pH 7.4 as dilution and wash buffers. Bound monovalent HA Fc fusions or anti-NIP IgG1 variants were detected by an anti-human IgG Fc specific AP-conjugated antibody from goat (Sigma-Aldrich, A9544) (diluted 1:5000 in PBS/T/S) and visualized by addition of p-nitrophenylphosphate (Merck) (10g/mL in diethanolamine buffer). Absorbance values were recorded at 405nm using a Sunrise plate-reader (TECAN).

The experimentswere performed at the Oslo University Hospital animal facility. Female homozygous B6.Cg-Fcgrttm1Dcr Tg(FcGRT)32Dcr/DcrJ mice (Tg32) aged 810 weeks (The Jackson Laboratory) (5 mice per group, 4 mice in the NaCl group) were used. Female mice were used due to housing considerations. At the day of vaccination, mice were anesthetized intraperitoneally with ZRF cocktail. When sedated, 10L of the vaccine mixtures was given in each nostril followed by inbreath while lying on their backs. Specifically, each mouse was given 20g CpG ODN 1826 Vaccigrade (Invivogen) mixed with 1.73g HA(H1N1)-Fc (WT or REW). After 3 weeks, each mouse was vaccinated in the same manner as above with 10% of the Fc fusions mixed with 20g CpG. The mice were challenged with a deadly dose (5x Lethal Dose 50) of influenza A H1N1 (A/Puerto Rico/8/1934 (PR8)) after 6 weeks. Specifically, mice were anesthetized and given virus i.n. as above. Weight loss was monitored daily or every second day after infection and the endpoint was set at 20% weight loss. If the endpoint was reached, mice were terminated by cervical dislocation or CO2 gas chamber. Plasma samples wereharvested by saphenous vein puncture 2 weeks after the boost to measure HA antibody titers. The experiments were approved by the Norwegian Food Safety Authority.

96-well EIA/RIA plates were coated with recombinant HA (H1N1 A/PR8) protein (1g/mL/100L per well) (Sino Biological). The wells were blocked with PBS/S and washed before addition of plasma samples diluted 1:1, 1:2, 1:4, 1:8, 1:16, 1:32, and 1:64 for 1h at RT. Bound HA specific mouse antibodies were detected using an AP-conjugated anti-mouse IgG Fc specific antibody from goat (Sigma-Aldrich, A9544) (diluted 1:4000 in PBS/T/S) and visualized by addition of AP-substrate (10g/mL in diethanolamine buffer). Absorbance values were measured at 405nm using a Sunrise Spectrophotometer (TECAN). PBS/S/T (pH 7.4) or PBS/T were used as dilution and washing buffer, respectively. Antibody titer was determined as the highest dilution factor for each mouse with a higher OD value than background, where the background is the mean absorbance of mice given NaCl plus 5x the standard error of the mean of the same observations.

An established ex vivo placental model was used39,71,72. Here, 10 g/ml of NIP-specific IgG1 variants mixed 1:1 with infliximab (anti-TNF, IgG1; Remicade) in a total volume of 100ml was added to the maternal perfusion reservoirs. Samples from the fetal reservoir were collected before the test proteins were added (0min) and from 2min after adding the antibodies (perfusion start) followed by the time points 60, 120, 180, 210, 240, 270, 300, 330 and 360min. Placentas from uncomplicated pregnancies resulting in vaginal birth or caesarean section were donated by women giving birth at Copenhagen University Hospital. To minimize variation, mothers who smoked, had diabetes or other pregnancy complications were excluded from the study. Only term placentas were included, and the project was approved by the ethical committees in the Communities of Copenhagen and Frederiksberg and the Danish Data Protection Agency. Informed consent was obtained in accordance with the Declaration of Helsinki.

Quantification offrom cellular experiments, in vivo studies and the ex vivo placental perfusion model was performed using ELISA. Recombinant mAb variants were captured on either BSA-NIP(25) (1g/mL/100L per well) (Biosearch Technologies), TNF- (1g/mL/100L per well) (Peprotech), recombinantly produced SARS-CoV-2 receptor binding domain (RBD) (Sino Biologics) (1g/mL/100L per well) or anti-human IgG Fc (Sigma-Aldrich, I2136) (1g/mL/100L per well). Captured mAb variants were detected by an APconjugated polyclonal anti-human IgG Fc specific Ab from goat (Sigma-Aldrich, A9544) (diluted 1:5000 in PBS/T/S) and visualized by addition of p-nitrophenylphosphate (Sigma-Aldrich) (10g/mL in diethanolamine buffer). Absorbance was recorded at 405nm using a Sunrise plate-reader (TECAN). PBS/S/T (pH 7.4) or PBS/T was used as dilution and washing buffers, respectively.

ELISAswere performed by capturing anti-NIP IgG1 variants (3000.023.4ng/mL) on BSA-NIP(16) (diluted to 1g/mL in PBS) (Biosearch technologies) coated in 96-well EIA/RIA plates (CorningCostar) or by randomly coating NIP IgG1 mAbs directly in wells (10g/mL). For antigen density experiments BSA-NIP(3), BSA-NIP(16) or BSA-NIP(80) (Norwegian Institute of Public Health) were coated at BSA concentrations of 2.5g/mL, 0.5g/mL and 0.1g/mL, respectively, to obtain different surface densities with the same number of total NIP molecules. NHS (diluted 1:200 in veronal buffer) (Complement Technologies), pure human C1q (diluted to 0.366g/mL in veronal buffer) (Complement Technologies) or pure mouse C1q (diluted to 0.366g/mL in veronal buffer) (Prospecbio) was added and the plates incubated at 37C for 30min. Bound C1q or deposited C3, C4 or C5 was detected using specific primary antibodies from rabbit (all from Dako/Agilent, A0135, A0062, A0065, A0056) (diluted 1:5000 in PBS/T/S) and a secondary HRP-conjugated anti-rabbit IgG antibody from donkey (Cytiva Life Sciences, NA935) (diluted 1:10.000 in PBS/T/S). Bound mouse C1q was detected by polyclonal anti-mouse C1q serum from goat (Creative Biolabs, CTA-P-023) (diluted 1:2000 in PBS/T/S) followed by detection using an anti-goat IgG AP-conjugated antibody from rabbit (Merck). Binding was visualized by addition of TMB substrate (Calbiochem) or AP-substrate (Merck) (10g/mL in diethanolamine buffer). The HRP reaction was terminated by adding 1M HCl. Detection of deposited TCC (C5bC9) was done using a biotinylated mouse monoclonal antibody specific for a C9 neoepitope exposed upon C5b binding (Diatec Monoclonals, DIA 011-01) (diluted 1:5000 in PBS/T/S) together with AP-conjugated streptavidin (diluted 1:5000 in PBS/T/S) (Cytiva Life Sciences) and visualized by addition of AP substrate (Sigma-Aldrich). For the spA competition ELISA, a 5-molar excess of spA (Sigma-Aldrich) was allowed to bind captured NIP IgG1 variants before addition of pure C1q. Absorbance values were recorded at either 450 (HRP) or 405 (AP) nm using a Sunrise plate-reader (TECAN). PBS/S was used as blocking buffer while PBS/S/T was used as dilution and washing buffers.

Solution phase complement assayswere performed by incubating anti-NIP IgG1 WT, REW, RGY, and PGLALA in NHS (Complement Technologies) at a concentration of 100g/mL for 1h at 37C. IgG complex formation and C4d concentrations were determined using the MicroVue CIC EIA and MicroVue C4d EIA kits (Quidel) following the manufacturer instructions.

Homozygous human FcRn transgenic mice (B6.Cg-Fcgrttm1Dcr Tg(FcGRT)32Dcr/DcrJ) (Tg32) (The Jackson Laboratory) were injected i.v. with 10mg/kg anti-NIP IgG1 WT, REW, and RGY variants (3 mice per group, age 810 weeks, 2 female and 1 male per group) followed by terminal bleeding and plasmacollection 24hpost-administration. In addition, plasma from 3 non-treated mice were collected. The plasma levels of anti-NIP IgG1 WT, REW and RGY were quantified by the anti-human Fc ELISA described above by interpolation to 12-point standard curves of each individual protein (10000.0056g/mL). Binding between anti-NIP IgGs and mouse C1q in the plasma samples was measured by coating anti-mouse C1q serum from goat (Creative Biolabs, CTA-P-023) (diluted 1:500 in PBS) in 96-well EIA/RIA plates followed by blocking for 2h ar RT. Then human IgG concentration normalized plasma samples were added and incubated for 1h at RT. Bound human IgG1 variants were detected by a goat anti-human IgG Fc specific AP-conjugated antibody (Sigma-Aldrich A5944) (diluted 1:5000 in PBS), and binding visualized by addition of AP-substrate (10g/mL in diethanolamine buffer) (Sigma-Aldrich). Absorbance values were recorded at 405nm using a Sunrise Spectrophotometer (TECAN). PBS/S was used as blocking buffer, PBS/S/T was used as dilution buffer and PBS/T was used as wash buffer in between each step. Levels of mouse C3a in human IgG concentration normalized plasma samples were measured using TECO Mouse C3a assay kit (Quidel) following the manufacturers instructions.

Raji, WSU-NHL, DOHH2, SU-DHL-4 cells were washed in 10mL HBSS and resuspended to 1.0107 cells/mL in HBSS before stained with 1M Calcein-AM (Merck) for 20min at RT. The cells were then washed twice with 10mL HBSS and diluted in HBSS to a density of 1.0106 cells/mL before 50 L was added to 96-well V-bottom plates (5.0104 cells/well) together with 25L NHS (Complement Technologies) (25% final concentration) and 25l anti-CD20 IgG1 variants (0.74g/mL final concentration). 50L RIPA buffer (ThermoFisher) instead of antibodies and NHS was used to determine maximum lysis of the target cells. Control samples containing 25L HBSS+50 L cells and 25L NHS were included to account for background. The plates were incubated at 37C/5% CO2 for 1h before centrifuged at 1314g for 5min. 50l supernatant was then transferred to black clear bottom optical 96-well Viewplates (Perkin Elmer) and fluorescent intensity was determined at 485nm excitation/510nm emission using an Envision plate reader (Perkin Elmer). Percent antibody-mediated lysis was calculated relative to the max lysis control.

GFP-expressing Newman spa/sbi or Newman WT (7.5105 CFU) was incubated with human monoclonal anti-WTA (4497) IgG1 WT, REW, or PGLALA, and 1% IgG/IgM serum in RPMI+0.05% HSA (RPMI-HSA), for 15min at 37C with shaking (700rpm), in a round-bottom microplate. Bacteria were then incubated with freshly isolated human neutrophils (7.5104), that were purified from blood of healthy donors by the Ficoll/Histopaque density gradient method73, for another 15min at 37C with shaking. All samples were fixed with 1% paraformaldehyde in RPMI-HSA. The binding/internalization of GFP bacteria to the neutrophils was detected using flow cytometry (BD FACSVerse), and data were analyzed based on forward/side scatter gating of neutrophils using FlowJo software. The use of human neutrophils was approved by Ethics Committee NedMec under informed consent from healthy donors.

Human neutrophils were freshly isolated from healthy donor blood using the Ficoll-Histopaque gradient method74. Phagocytosis assay was performed in a round-bottom 96-well plate and neutrophil-associated fluorescent bacteria were analyzed by flow cytometry. FITC-labeled S. pneumoniae were opsonized by pre-incubation with 2-fold serial dilutions of the antibodies in IgG/IgM-depleted serum75 as complement source, in RPMI-HSA for 20min at 37C. Subsequently, neutrophils were added in a 1:10 cell to bacteria ratio and phagocytosis was allowed for 15 (S. aureus) or 30min (S. pneumoniae) at 37C on a shaker (650rpm). Ice-cold 1% PFA in RPMI-HSA was added to stop the reaction. Samples were measured by flow cytometry, and % of positive cells and mean fluorescence values were determined for gated neutrophils74. The use of human neutrophils was approved by Ethics Committee NedMec under informed consent from healthy donors.

Bacteria harvested from ON cultures were re-passaged on chocolate agar, grown for 6h, and suspended in HBSS containing 0.15mM CaCl2 and 1mM MgCl2 (HBSS++). About 2.000 CFU of suspended bacteria were incubated with human complement (IgG and IgM depleted normal human serum; Pel-Freez) and titrated amounts of 2C7 IgG variants. The final reaction volumes were maintained at 150L. Aliquots of 25L of the reaction mixtures were plated onto chocolate agar in duplicates at the beginning of the assay (t0) and again after incubation at 37C for 30min (t30). Survival was calculated as the number of CFUs at t30 relative to t0. Binding of Humab 2C7 variants to the surface of N. gonorrhoeae was measured performed by flow cytometry76.

Anti-NIP IgG1 variants (2g/mL in PBS) were captured on BSA-NIP(25) (1 g/mL in PBS) (Biosearch Technologies) coated in 96-well EIA/RIA plates (CorningCostar) for 1h at RT. Biotinylated human FcRI, FcRIIa-H131, FcRIIa-R131, FcRIIb, FcRIIIA-V158 and FcRIIA-F158 (10g/mL) (Sino Biological) were then added and incubated for 1h at RT. Bound receptors were detected with streptavidin-AP (diluted 1:500 in PBS/T/S) (Roche Diagnostics). Biotinylated FcRIIIb (1 g/mL in PBS) was pre-incubated with streptavidin-AP to increase the detection sensitivity. The 405nm absorption spectrum was recorded using a Sunrise plate-reader (TECAN). PBS/S (pH 7.4) was used as blocking buffer while PBS/S/T (pH 7.4) and PBS/T was used as dilution and washing buffers, respectively. A Biacore T200 instrument (Cytiva Life Sciences) was used to obtain SPR binding curves of NIP IgG1 variants to biotinylated human FcRI, FcRIIa-H131, FcRIIa-R131, FcRIIb, FcRIIIA-V158 and FcRIIA-F158 (Sino Biological). The receptors were captured on Series S SA chips at 200 RU and binding to the low affinity FcRs recorded by single injections of 100nM anti-NIP IgG1 variants at a flow rate of 10 L/min and a contact time of 60s. For high affinity FcRI, 50nM anti-NIP IgG1 was injected at a flow rate of 40L/min with a contact time of 30s. The maximum binding responses were normalized to 100 RU using the BIAevaluation software to allow overlay of the sensorgrams.

ADCCwas performed byan [51Cr] release assay50. MNCs, anti-CD20 IgG1 variants (0.01100nM) and medium were added to round-bottom microtiter plates (Nunc, Rochester, NY, USA). Assays were started by addition of effector and target cells at E:T ratio 40:1 (200000:5000). After 3h at 37C, [51Cr] release from 5 parallel samples was measured. Percentage of [51Cr] release was calculated using the formula: % lysis=(experimental cpm basal cpm)/(maximal cpm basal cpm) 100; maximal [51Cr] release was determined by adding Triton-X (2% final concentration) to target cells, and basal release measured in the absence of antibodies. The use of human MNCs were approved by the Ethics Committee of Kiel University under informed consent from healthy donors.

Monocyte derived macrophages were generated by letting them attach to the cell culture flask for 30min at 37C in monocyte-attachment medium (PromoCell). The monocyte attachment medium was then removed, and the cell culture flasks were washed with 1xPBS before the monocytes were cultured in X-VIVO 15 media (Lonza) with 25g/mL streptomycin and 25U/mL penicillin (ThermoFisher). For macrophage generation 50ng/mL M-CSF (PeproTech) was added every 3 days for at least 9 days. Real time, automated ADCP was measured by fluorescence microscopy using Incucyte (Sartorius). Target cells were labeled with 0.5g/mL pHrodo dye (ThermoFisher) for 1h at room temperature. M-CSF derived macrophages were added as effector cells using an effector/target cell ratio of 1:1 (40.000:40.000). The assay was started by adding anti-CD20 mAb2 antibody variants (100nM). Then, ADCP was measured for 10h every 10min at 37C. ADCP was determined as red counts per image, which correlate to the number of tumor cells engulfed by the macrophages. Analysis was performed using Top-Hat segmentation, 2 red calibrated units (RCU) as threshold and 20m radius. The minimum intensity was set to 88.5. Under these conditions control samples (pHrodo cells only) started at 10 counts at time point 0. The use of human MNCs was approved by the Ethics Committee of Kiel University under informed consent from healthy donors.

An established CDC 51Cr assay was used77. Target cells per condition were labeled with 200Ci [51Cr] for 2h. To the Raji target cells (1.0104 per condition), 25% v/v freshly drawn human serum was added as source of complement in the presence of anti-CD20 IgG1 variants (0.01100nM). The percentage of cellular cytotoxicity was calculated using the same formula as for 51Cr ADCC assays. The use of human MNCs were approved by the Ethics Committee of Kiel University under informed consent from healthy donors.

Protein samples (2g) were analyzed on 12% Bis-Tris Bolt SDS-PAGE gels (ThermoFisher) with Bolt MES SDS running buffer (ThermoFisher). ASpectra Multicolor Broad Range Protein Ladder (ThermoFisher) was used as standard, and the gels were stained with Comassie Brilliant Blue (BioRad).

Analytical SECwas performed by applying 20g of each anti-NIP IgG variant to a Superdex 200 Increase 3.2/300 analytical SEC column (Cytiva Lifesciences) at a flow rate of 0.05mL/min using an AKTA FPLC instrument (Cytiva Life sciences). Data was normalized to relative fluorescence for clarity.

DSFwas performed either by a dye-based method using a Lightcycler RT-PCR instrument (Roche) or label-free using a Prometheus NT.48 nanoDSF instrument (Nanotemper Technologies GmbH). For dye-based DSF, SYPO Orange (Sigma-Aldrich) was used at a dilution of 1:1000 with a protein concentration 0.1mg/mL in a final volume of 25L. All samples were run in triplicates in 96-well Lightcycler 480 multiwell plates. The peaks of excitation and emission filters were used and the instrument programed to raise the temperature from 25C to 95C after a stabilization time of 10min at 25C. Data was collected every 0.5C. Data transformation and analysis were performed using a DSF analysis protocol78. For label-free DSF, 1mg/mL samples were drawn into capillaries in triplicates. The instrument was set to gradually increase the temperature(2 C/min) from 25C to 95C. The melting temperature (Tm C) in which half of the proteins were unfolded was determined by deducing the first derivative in the PR. ThermoControl software.

Recombinantly produced IgG1 WT and REW Fc fragments were coated (1g/mL/100uL per well) in 96-well EIA/RIA plates (CorningCostar) for 1h at RT. Then Rh+ human serum (Lee BioSolutions) was diluted 1:11:106-fold in PBS, added to the plates and incubated for 1h at RT. NHS was added at 1:1 dilution as a negative control. The plates were washed 4 times with PBS/T/S. Serum antibodies bound to the coated Fc fragments were detected using a pan anti-human IgG light chain antibody from rabbit (ReMab Biosciences, 32-1031-00) (diluted 1:3000 in PBS/T/S) and visualized by addition of anti-rabbit IgG-HRP (Cytiva Life Sciences, NA935) (diluted 1:3000 in PBS/T/S). The coating efficacy of WT and REW Fc fragments were controlled using an AP conjugated polyclonal anti-human IgG (Fc specific) antibody from goat (Sigma-Aldrich, A5944) (diluted 1:5000 in PBS/T/S). The 450nm or 405nm absorbance values were recorded using a Sunrise plate reader (Tecan).

was performed using a human FcRn retention column and a KTA Avant 25 instrument (Cytiva Life Sciences). 100g NIP IgG1 variants were injected over the column and allowed to bind FcRn at pH 6.0 before being subjected to an increasingly more basic pH gradient (pH 6.0 to 8.8) over 110min by mixing of two eluent buffers (20mM MES sodium salt, 140mM NaCl, pH 6.0 and 20mM Tris/HCl, 140 NaCl, pH 8.8). The pH was continuously recorded by a pH detector (Cytiva Life Sciences).

Titrated amounts of NIP IgG1 WT or REW (1000.00.78ng/mL) was captured on BSA-NIP(25) (diluted to 1g/mL in PBS) coated in 96-well EIA/RIA plates (CorningCostar) pre-blocked with PBS/S. Then AP conjugated Protein A from S. aureus(Sigma-Aldrich) was diluted 1:5000 in PBS/T/S before visualization of binding by addition of AP substrate (diluted to 10g/mL in diethanolamine). The 405nm absorption values were recorded using a Sunrise spectrophotometer (TECAN).

Liquid chromatography tandem mass spectrometrywas performed by mixing 50l of each IgG1 variant (1mg/mL) with 1g trypsin dissolved in 100l 50mM ammonium bicarbonate (pH 7.8) and incubated ON at 37C. Peptides were isolated by collecting the flow-through from centrifugal filters and transferred to Eppendorf tubes before drying using SpeedVac (HetMaxi dry). Dried samples were dissolved in 20l 0.1% formic acid, sonicated for 30s and centrifuged for 10min at 16,100g. 10l of each samples was transferred to new vials, and reverse phase (C18) nano outline liquid chromatography-tandem mass spectrometry (LC-MS/MS) analysis of proteolytic peptides was performed using a system of two Agilent 1200 HPLC binary pumps (nano and capillary) with an autosampler, column heater and integrated switching valve. The system was coupled via a nanoelectronspray ion spource to an LTQ Orbitrap mass spectrometer (ThermoFisher). For the analysis 6l peptide solution was injected into the 50.3mm extraction column filled with Zorbax 300SB-C18 of 5m particle size (Agilent Technologies). After washing for 5min with 0.1% formic acid (v/v) and 3% acetonitrile (v/v) at a flow rate of 10l/min, the integrated switching valve was activated and peptides were eluted in the back-flush mode from the extraction column onto a 1500.075mm C18, 2m resin column (GlycoproSIL C18-80, Glycopromass). The mobile phase consisted of acetonitrile and mass spectroscopy-grade water both containing 0.1% formic acid. Chromatographic separation was achieved using a binary gradient from 5 to 55% of acetonitrile in water for 1h with a flow rate of 0.2l/min. Mass spectra were acquired in the positive ion mode applying a data-dependent automatic switch between survey scan and MS/MS acquisition. Peptide samples were analyzed with a high-energy collisional dissociation (HCD) fragmentation method with normalized collision energy at 25 and 41, acquiring one Orbitrap survey scan in the mass range of m/z 3002000 followed by MS/MS of the three most intense ions in the Orbitrap (R7500). The target value in the LTQ-Orbitrap was 1 million for survey scan at a resolution of 30,000 at m/z 400 using lock masses for recalibration to improve the mass accuracy of precursor ions. Ion selection threshold was 500 counts. Selected sequenced ions were dynamically excluded for 180s. Data analysis was performed on Xcalibur v2.0. MS/MS spectra for all glycopeptides and thesewere extracted by oxonium ion search; 204.086 (N-acetylhexosamine) and 366.1388 (N-acetylhexosamine-hexose) were used. HCD fragmentation with normalized collision energy at 25 was used to detect the glycans, and the peptide mass was detected for the IgG glycopeptides. Extracted ion chromatogram for target glycopeptides (EEQYNSTYR and the miscleaved TKPREEQYNSTYR with all different glycan masses) were extracted with 10ppm accuracy and MS/MS spectra were manually verified. HCD fragmentation with normalized collision energy at 41 was used to detect the peptide sequence and to verify that the peptide mass corresponded to the correct peptide sequence. The area under the curve for all extracted glycopeptides was calculated and the percentage ratio for each glycoform was determined.

Prediction of T-cell epitopes was performed using NetMHC4.179 with 9mer peptides against representative human HLA supertypes using the default rank thresholds for strong (0.5) and weak (2.0) binders. B-cell epitope prediction was performed using the Immune Database analysis resource antibody epitope prediction tool80 with a threshold of 0.5.

Figures and statistical analyses were prepared using GraphPad Prism (GraphPad Software).

Further information on research design is available in theNature Portfolio Reporting Summary linked to this article.

Read more:

Human IgG Fc-engineering for enhanced plasma half-life, mucosal distribution and killing of cancer cells and bacteria - Nature.com

Read More..

Gill S, Olson JA, Negrin RS. Natural killer cells in allogeneic transplantation: effect on engraftment, graft- versus-tumor, and graft-versus-host responses. Biol Blood Marrow Transplant. 2009;15:76576.

Article CAS PubMed Central PubMed Google Scholar

Freud AG, Mundy-Bosse BL, Yu J, Caligiuri MA. The broad spectrum of human natural killer cell diversity. Immunity. 2017;47:82033.

Article CAS PubMed Central PubMed Google Scholar

Dogra P, Rancan C, Ma W, Toth M, Senda T, Carpenter DJ, et al. Tissue determinants of human NK cell development, function, and residence. Cell. 2020;180:749763.e13.

Article CAS PubMed Central PubMed Google Scholar

Buckle I, Johnson A, Rojas IL, Weinert V, Sester DP, Radford K, et al. High dimensional analysis reveals distinct NK cell subsets but conserved response to stimulation in umbilical cord blood and adult peripheral blood. Eur J Immunol. 2023;53:2250118.

Article CAS Google Scholar

Wang W. NK cell-mediated antibody-dependent cellular cytotoxicity in cancer immunotherapy. Front Immunol. 2015;6. https://doi.org/10.3389/fimmu.2015.00368.

Gong JH, Maki G, Klingemann HG. Characterization of a human cell line (NK-92) with phenotypical and functional characteristics of activated natural killer cells. Leukemia. 1994;8:6528.

CAS PubMed Google Scholar

Jochems C, Hodge JW, Fantini M, Fujii R, Morillon YM, Greiner JW, et al. An NK cell line (haNK) expressing high levels of granzyme and engineered to express the high affinity CD16 allele. Oncotarget. 2016;7:8635973.

Article PubMed Central PubMed Google Scholar

Tonn T, Schwabe D, Klingemann HG, Becker S, Esser R, Koehl U, et al. Treatment of patients with advanced cancer with the natural killer cell line NK-92. Cytotherapy. 2013;15:156370.

Article CAS PubMed Google Scholar

Liu S, Galat V, Galat Y, Lee YKA, Wainwright D, Wu J. NK cell-based cancer immunotherapy: from basic biology to clinical development. J Hematol Oncol. 2021;14:7.

Article CAS PubMed Central PubMed Google Scholar

Bryceson YT, March ME, Ljunggren H-G, Long EO. Synergy among receptors on resting NK cells for the activation of natural cytotoxicity and cytokine secretion. Blood. 2006;107:15966.

Article CAS PubMed Central PubMed Google Scholar

Szmania S, Lapteva N, Garg T, Greenway A, Lingo J, Nair B, et al. Ex Vivoexpanded natural killer cells demonstrate robust proliferation in vivo in high-risk relapsed multiple myeloma patients. J Immunother. 2015;38:2436.

Article CAS PubMed Central PubMed Google Scholar

Berrien-Elliott MM, Becker-Hapak M, Cashen AF, Jacobs M, Wong P, Foster M, et al. Systemic IL-15 promotes allogeneic cell rejection in patients treated with natural killer cell adoptive therapy. Blood. 2022;139:117783.

Article CAS PubMed Central PubMed Google Scholar

Bachanova V, Cooley S, Defor TE, Verneris MR, Zhang B, McKenna DH, et al. Clearance of acute myeloid leukemia by haploidentical natural killer cells is improved using IL-2 diphtheria toxin fusion protein. Blood. 2014;123:385563.

Article CAS PubMed Central PubMed Google Scholar

Miller JS, Soignier Y, Panoskaltsis-Mortari A, McNearney SA, Yun GH, Fautsch SK, et al. Successful adoptive transfer and in vivo expansion of human haploidentical NK cells in patients with cancer. Blood. 2005;105:30517.

Article CAS PubMed Google Scholar

Heuser M, Tschan-Plessl A, Thol F, Schwarzer A, Kloos A, Kattre N, et al. Allogeneic, CD34 +, umbilical cordblood-derived NK cell adoptive immunotherapy for the treatment of acute myeloid leukemia patients with measurable residual disease. Blood. 2021;138:17451745.

Article Google Scholar

Choi I, Yoon SR, Park S-Y, Kim H, Jung S-J, Kang Y-L, et al. Donor-derived natural killer cell infusion after human leukocyte antigenhaploidentical hematopoietic cell transplantation in patients with refractory acute leukemia. Biol Blood Marrow Transplant. 2016;22:206576.

Article CAS PubMed Google Scholar

Bachanova V, Maakaron JE, Cichocki F, McKenna DH, Cao Q, DeFor TE, et al. Gda-201, a novel metabolically enhanced allogeneic natural killer (NK) cell product yields high remission rates in patients with relapsed/refractory non-Hodgkin lymphoma (NHL): 2-year survival and correlation with cytokine IL7. Blood. 2021;138:38543854.

Article Google Scholar

Nangia C, Soon-Shiong P, Rabizadeh S, Lee JH, Sender L, Jones F, et al. Complete responses in patients with second-line or greater metastatic triple negative breast cancer (TNBC) following first-in-human immunotherapy combining NK and T cell activation with off-the-shelf high-affinity CD16 NK cell line (haNK). Ann Oncol. 2019;30:v130.

Article Google Scholar

Fate Therapeutics. Fate therapeutics announces encouraging dose-escalation clinical data of FATE-NK100 and provides regulatory update on landmark IND appli cation for FT500 [Press Release]. 2018. https://ir.fatetherapeutics.com/press-releases?field_nir_news_date_value%5Bmin%5D.

Depil S, Duchateau P, Grupp SA, Mufti G, Poirot L. Off-the-shelf allogeneic CAR T cells: development and challenges. Nat Rev Drug Discov. 2020;19:18599.

Article CAS PubMed Google Scholar

Yoon DH, Koh Y, Park H, Hwang YK, Kim WS. A phase 1 study of the combination of MG4101, ex vivo -expanded allogeneic NK cells and rituximab for relapsed or refractory Non-Hodgkin Lymphoma. Blood. 2020;136:1415.

Article Google Scholar

Bachanova V, Sarhan D, DeFor TE, Cooley S, Panoskaltsis-Mortari A, Blazar BR, et al. Haploidentical natural killer cells induce remissions in non-Hodgkin lymphoma patients with low levels of immune-suppressor cells. Cancer Immunol Immunother. 2018;67:48394.

Article CAS PubMed Google Scholar

Liang S, Lin M, Niu L, Xu K, Wang X, Liang Y, et al. Cetuximab combined with natural killer cells therapy: an alternative to chemoradiotherapy for patients with advanced non-small cell lung cancer (NSCLC). Am J Cancer Res. 2018;8:87991.

CAS PubMed Central PubMed Google Scholar

Lin M, Luo H, Liang S, Chen J, Liu A, Niu L, et al. Pembrolizumab plus allogeneic NK cells in advanced non-small cell lung cancer patients. J Clin Invest. 2020;130:25609.

Article CAS PubMed Central PubMed Google Scholar

Chung YY, Park SW, Im J-M, Yoo D-K, Cheon H-C, Kim J-E, et al. Abstract CT171: combinatorial allogeneic NK cell therapy with Pembrolizumab for cholangiocarcinoma; interim report of open label Phase1/2a study. Cancer Res. 2021;81:CT171CT171.

Article Google Scholar

Adotevi O, Godet Y, Galaine J, Lakkis Z, Idirene I, Certoux JM, et al. In situ delivery of allogeneic natural killer cell (NK) combined with Cetuximab in liver metastases of gastrointestinal carcinoma: a phase I clinical trial. OncoImmunology. 2018;7:e1424673.

Article CAS PubMed Central PubMed Google Scholar

Federico SM, McCarville MB, Shulkin BL, Sondel PM, Hank JA, Hutson P, et al. A pilot trial of humanized anti-GD2 monoclonal antibody (hu14.18K322A) with chemotherapy and natural killer cells in children with recurrent/refractory neuroblastoma. Clin Cancer Res. 2017;23:64419.

Article CAS PubMed Central PubMed Google Scholar

Talleur AC, Triplett BM, Federico S, Mamcarz E, Janssen W, Wu J, et al. Consolidation therapy for newly diagnosed pediatric patients with high-risk neuroblastoma using Busulfan/Melphalan, autologous hematopoietic cell transplantation, anti-GD2 antibody, granulocyte-macrophage colonystimulating factor, interleukin-2, and haploidentical natural killer cells. Biol Blood Marrow Transplant. 2017;23:19107.

Article CAS PubMed Central PubMed Google Scholar

Schmidt P, Raftery MJ, Pecher G. Engineering NK cells for CAR therapyrecent advances in gene transfer methodology. Front Immunol. 2021;11:611163.

Article PubMed Central PubMed Google Scholar

Gong Y, Klein Wolterink RGJ, Janssen I, Groot AJ, Bos GMJ, Germeraad WTV. Rosuvastatin enhances VSV-G lentiviral transduction of NK cells via upregulation of the low-density lipoprotein receptor. Mol Ther Methods Clin Dev. 2020;17:63446.

Article CAS PubMed Central PubMed Google Scholar

Mller S, Bexte T, Gebel V, Kalensee F, Stolzenberg E, Hartmann J, et al. High cytotoxic efficiency of lentivirally and alpharetrovirally engineered CD19-specific chimeric antigen receptor natural killer cells against acute lymphoblastic leukemia. Front Immunol. 2020;10:3123.

Article PubMed Central PubMed Google Scholar

Sengsayadeth S, Savani BN, Oluwole O, Dholaria B. Overview of approved CART therapies, ongoing clinical trials, and its impact on clinical practice. eJHaem. 2022;3:610.

Article PubMed Google Scholar

Zhang X, Guo Y, Ji Y, Gao Y, Zhang M, Liu Y, et al. Cytokine release syndrome after modified CAR-NK therapy in an advanced non-small cell lung cancer patient: a case report. Cell Transpl. 2022;31:096368972210942.

Article Google Scholar

Klingemann H. Are natural killer cells superior CAR drivers? OncoImmunology. 2014;3:e28147.

Article PubMed Central PubMed Google Scholar

Tang X, Yang L, Li Z, Nalin AP, Dai H, Xu T, et al. First-in-man clinical trial of CAR NK-92 cells: safety test of CD33-CAR NK-92 cells in patients with relapsed and refractory acute myeloid leukemia. Am J Cancer Res. 2018;8:10839.

CAS PubMed Central PubMed Google Scholar

Fate Therapeutics. Fate Therapeutics Showcases Positive Interim Phase 1 Data from FT596 Of-the-shelf, iPSC-derived CAR NK Cell Program for Relapsed/Refrac tory B-cell Lymphoma at 2021 ASH Annual Meeting [Press Release]. 2021. https://ir.fatetherapeutics.com/press-releases?field_nir_news_date_value%5Bmin%5D=2021.

Bachanova V, Cayci Z, Lewis D, Maakaron JE, Janakiram M, Bartz A, et al. Initial clinical activity of FT596, a first-in-class, multi-antigen targeted, off-the-shelf, iPSC-derived CD19 CAR NK cell therapy in relapsed/refractory B-cell lymphoma. Blood. 2020;136:88.

Article Google Scholar

Liu E, Marin D, Banerjee P, Macapinlac HA, Thompson P, Basar R, et al. Use of CAR-transduced natural killer cells in CD19-positive lymphoid tumors. N Engl J Med. 2020;382:54553.

Article CAS PubMed Central PubMed Google Scholar

Marin D, Li Y, Basar R, Rafei H, Daher M, Dou J et al. Safety, efficacy and determinants of response of allogeneic CD19-specific CAR-NK cells in CD19+ B cell tumors: a phase 1/2 trial. Nat Med. 2024. https://doi.org/10.1038/s41591-023-02785-8.

Jiang H, Zhang W, Shang P, Zhang H, Fu W, Ye F, et al. Transfection of chimeric anti-CD138 gene enhances natural killer cell activation and killing of multiple myeloma cells. Mol Oncol. 2014;8:297310.

Article CAS PubMed Google Scholar

Romanski A, Uherek C, Bug G, Seifried E, Klingemann H, Wels WS, et al. CD 19 CAR engineered NK 92 cells are sufficient to overcome NK cell resistance in Bcell malignancies. J Cell Mol Med. 2016;20:128794.

Article CAS PubMed Central PubMed Google Scholar

Oelsner S, Waldmann A, Billmeier A, Rder J, Lindner A, Ullrich E, et al. Genetically engineered CAR NK cells display selective cytotoxicity against FLT3positive BALL and inhibit in vivo leukemia growth. Int J Cancer. 2019;145:193545.

Chu J, Deng Y, Benson DM, He S, Hughes T, Zhang J, et al. CS1-specific chimeric antigen receptor (CAR)-engineered natural killer cells enhance in vitro and in vivo antitumor activity against human multiple myeloma. Leukemia. 2014;28:91727.

Article CAS PubMed Google Scholar

Mansour AG, Teng K-Y, Li Z, Zhu Z, Chen H, Tian L, et al. Off-the-shelf CARengineered natural killer cells targeting FLT3 enhance killing of acute myeloid leukemia. Blood Adv. 2023;7:622539.

Article CAS PubMed Central PubMed Google Scholar

Chen Y, You F, Jiang L, Li J, Zhu X, Bao Y, et al. Gene-modified NK-92MI cells expressing a chimeric CD16-BB- or CD64-BB- receptor exhibit enhanced cancer-killing ability in combination with therapeutic antibody. Oncotarget. 2017;8:3712839.

Article PubMed Central PubMed Google Scholar

Chu Y, Hochberg J, Yahr A, Ayello J, van de Ven C, Barth M, et al. Targeting CD20+ aggressive B-cell NonHodgkin lymphoma by anti-CD20 CAR mRNA-modified expanded natural killer cells in vitro and in NSG mice. Cancer Immunol Res. 2015;3:33344.

Article CAS PubMed Google Scholar

Caruso S, De Angelis B, Del Bufalo F, Ciccone R, Donsante S, Volpe G, et al. Safe and effective off-the-shelf immunotherapy based on CAR.CD123-NK cells for the treatment of acute myeloid leukaemia. J Hematol Oncol. 2022;15:163.

Article CAS PubMed Central PubMed Google Scholar

Klaihmon P, Kang X, Issaragrisil S, Luanpitpong S. Generation and functional characterization of anti-CD19 chimeric antigen receptor-natural killer cells from human induced pluripotent stem cells. IJMS. 2023;24:10508.

Article CAS PubMed Central PubMed Google Scholar

Gang M, Marin ND, Wong P, Neal CC, Marsala L, Foster M, et al. CAR-modified memory-like NK cells exhibit potent responses to NK-resistant lymphomas. Blood. 2020;136:230818.

Article PubMed Central PubMed Google Scholar

Albinger N, Bexte T, Buchinger L, Wendel P, Al-Ajami A, Gessner A, et al. CRISPR/Cas9 gene editing of immune checkpoint receptor NKG2A improves the efficacy of primary CD33-CAR-NK cells against AML. Blood. 2022;140:45589.

Article Google Scholar

Portillo AL, Hogg R, Poznanski SM, Rojas EA, Cashell NJ, Hammill JA, et al. Expanded human NK cells armed with CAR uncouple potent anti-tumor activity from off-tumor toxicity against solid tumors. iScience. 2021;24:102619.

Article ADS CAS PubMed Central PubMed Google Scholar

Ao X, Yang Y, Li W, Tan Y, Guo W, Ao L, et al. Anti-FR CAR-engineered NK-92 cells display potent cytotoxicity against FR-positive ovarian cancer. J Immunother. 2019;42:28496.

Article CAS PubMed Central PubMed Google Scholar

Liu WN, So WY, Harden SL, Fong SY, Wong MXY, Tan WWS, et al. Successful targeting of PD-1/PD-L1 with chimeric antigen receptor-natural killer cells and nivolumab in a humanized mouse cancer model. Sci Adv. 2022;8:eadd1187.

More:

Development of NK cell-based cancer immunotherapies through receptor engineering | Cellular & Molecular Immunology - Nature.com

Read More..

Spirax-Sarco Engineering (LON:SPX Get Free Report) had its price objective upped by research analysts at Berenberg Bank from GBX 8,800 ($111.69) to GBX 9,800 ($124.38) in a research report issued to clients and investors on Friday, Marketbeat.com reports. The firm currently has a hold rating on the stock. Berenberg Banks price target indicates a potential downside of 7.37% from the stocks previous close.

Several other equities research analysts have also issued reports on the stock. Shore Capital reissued a sell rating on shares of Spirax-Sarco Engineering in a research note on Monday, December 18th. Jefferies Financial Group raised shares of Spirax-Sarco Engineering to a hold rating and set a GBX 9,740 ($123.62) target price on the stock in a research note on Wednesday, January 31st. Finally, Numis Securities reissued a sell rating on shares of Spirax-Sarco Engineering in a research note on Thursday. Two investment analysts have rated the stock with a sell rating, two have issued a hold rating and one has assigned a buy rating to the companys stock. Based on data from MarketBeat, the company has an average rating of Hold and a consensus target price of 102.13 ($129.63).

Get Our Latest Research Report on Spirax-Sarco Engineering

(Get Free Report)

Spirax-Sarco Engineering plc provides engineered solutions for the users of industrial and commercial steam systems, electrical heating and temperature management systems, and pumps and fluid path technologies. The company offers industrial and commercial steam systems, including condensate management, controls, and thermal energy management products and solutions for heating, curing, cooking, drying, cleaning, sterilizing, space heating, humidifying, vacuum packing, and producing hot water; electrical process heating and temperature management solutions, such as industrial heaters and systems, heat tracing, and various component technologies for industrial processes; and peristaltic and niche pumps and associated fluid path technologies, including pumps, tubing, and specialty filling systems and products for single-use applications.

Receive News & Ratings for Spirax-Sarco Engineering Daily - Enter your email address below to receive a concise daily summary of the latest news and analysts' ratings for Spirax-Sarco Engineering and related companies with MarketBeat.com's FREE daily email newsletter.

Read more:

Spirax-Sarco Engineering (LON:SPX) PT Raised to GBX 9800 - Defense World

Read More..