Entity

Time filter

Source Type

Shanghai, China

WuXi PharmaTech is a research and development services company headquartered in Shanghai, with operations in China and the United States. The company uses an open-access technology platform designed to enable efficient development of medicines for customers. WuXi acts as a contract researcher for most of the world's largest pharmaceutical, biotech and medical device companies and many smaller companies. As of April 2013, the company had more than 1,600 customers. Wikipedia.


Guo Y.,WuXi AppTec | Shalaev E.,Allergan, Inc. | Smith S.,Allergan, Inc.
TrAC - Trends in Analytical Chemistry | Year: 2013

The acceptable stability of the drug formulations is one of the basic requirements for pharmaceutical development and commercialization. The increasing application of enabling delivery techniques poses even more challenge to the drug physical stability of pharmaceutical formulations. After a brief review of regulatory requirements and recent drug recalls due to physical instability, we discuss the physical stability of the solid-state drug in amorphous dispersions with focus on analytical techniques, amorphous molecular mobility, drug-excipient interaction, and the effect of water. © 2013 Elsevier Ltd. Source


News Article
Site: http://www.nature.com/nature/current_issue/

No statistical methods were used to determine sample size. Genomic DNA was isolated from Brevibacillus parabrevis ATCC 8185 (Cedarlane Laboratories) using a GenElute Bacterial Genomic DNA Kit (Sigma-Aldrich). Gene constructs comprising F and A domains (F–A) and all three domains (F–A–PCP) were amplified by PCR from the lgrA gene using the following primers, designed using sequence alignment with A and PCP domains of a known structure and the study of Marahiel and co-workers5, 6. FA_fwd: 5′-AATCATCCATGGGAAGAATACTATTCCTAACAACATTTATGAGCAAAG-3′; FA_rev: 5′-AATCATCTCGAGTTACGCATCGGCCTGCACGTCT-3′; FAT_fwd: 5′-TGACTACCATGGGGAGAATACTATTCCTAACAACATTTATGAGC-3′; FAT_rev: 5′-CGTTGAGCGGCCGCTTGCTCCGTAAGCAGACGTTT-3′. PCR product for F–A–PCP was digested using NcoI and NotI (New England Biolabs) and ligated into a pET21-derived vector containing an N-terminal octa-histidine tag with a tobacco etch virus (TEV) protease cleavage site. The PCR product for F–A–PCP was cloned between NcoI and NotI restriction sites into a pET21-derived vector containing an N-terminal TEV cleavable octa-histidine tag and a C-terminal TEV cleavable calmodulin binding peptide (CBP) tag. The F–A protein was expressed in Escherichia coli BL21 (DE3) cells. A 10 ml aliquot of overnight culture was used to inoculate 1 l of lysogeny broth (LB) medium supplemented with 350 μg ml−1 kanamycin. The culture was grown at 37 °C to an optical density (OD ) of 0.6, before inducing protein expression using 0.5 mM isopropyl β-d-1-thiogalactopyranoside (IPTG) and reducing the temperature to 16 °C for 18 h. Cells were collected by centrifugation at 4 °C and resuspended in nickel binding buffer (2 mM imidazole, 150 mM NaCl, 0.25 mM tris-(2-carboxyethyl)phosphine (TCEP), 50 mM Tris-HCl (pH 7.0)). The cells were lysed by sonication on ice and centrifuged for 30 min at 20,000g at 4 °C. Clarified lysate was loaded onto a HiTrap IMAC FF column (GE Healthcare). F–A protein was eluted using a gradient of 2–250 mM imidazole. Fractions containing F–A were pooled, diluted tenfold with ion exchange binding buffer (0.25 mM TCEP, 20 mM Tris, pH 8.0), loaded onto a HiTrap Q HP column and eluted using a gradient to 100% elution buffer (1 M NaCl, 0.25 mM TCEP, 20 mM Tris-HCl (pH 8.0)). The eluted protein was concentrated using a 10K MWCO Amicon Ultra-15 filtration unit (EMD Millipore) and subjected to gel filtration chromatography using a HiLoad 16/600 Superdex 200 column (GE Healthcare) equilibrated with S200 buffer (150 mM NaCl, 0.25 mM TCEP, 20 mM Tris (pH 7.0)). Protein purity was confirmed using SDS–PAGE and native PAGE. Pure F–A was concentrated in storage buffer (25% glycerol, 150 mM NaCl, 0.25 mM TCEP, 20 mM Tris (pH 7.0)), flash-frozen with liquid nitrogen and stored at −80 °C for later use. F–A–PCP was expressed in E. coli BL21 EntD-(DE3) cells using the same protocol as above. Cells were pelleted, resuspended in CBP binding buffer (25 mM Tris-HCl (pH 7.5), 150 mM NaCl, 2 mM imidazole (pH 8.0), 2 mM CaCl , 2 mM β-mercaptoethanol (βME) and 0.1 mM phenylmethanesulfonyl fluoride (PMSF)), sonicated and clarified by centrifugation for 30 min at 20,000g at 4 °C. Clarified lysate was loaded onto a 30 ml calmodulin sepharose 4B column (GE Healthcare). F–A–PCP was eluted with elution buffer (25 mM Tris-HCl (pH 7.5), 150 mM NaCl, 2 mM EGTA, 2 mM βME and 0.1 mM PMSF). Protein was dialysed against binding buffer for a minimum of 4 h before being loaded onto a 5 ml HiTrap IMAC FF column (GE Healthcare) charged with Ni2+ and equilibrated in nickel binding buffer. F–A–PCP was eluted using a 60 ml gradient of 0–250 mM imidazole. Fractions containing F–A–PCP were pooled and affinity tags were removed by cleavage with TEV protease at room temperature overnight using a 1:4 mg ratio of TEV to F–A–PCP. Cleaved F–A–PCP was passed back over the nickel and calmodulin affinity columns, with the flow-through collected, concentrated and applied to a HiLoad 16/600 Superdex 200 (GE Healthcare) in S200 buffer. Pure F–A–PCP was concentrated to 5.0 mg ml−1 in storage buffer, flash-frozen in liquid nitrogen and stored at −80 °C. Amino-coenzyme A (NH-CoA)23 was prepared enzymatically starting from amino-pantetheine (WuXi AppTec) using a previously published protocol31 with the following modifications: one-pot synthesis was carried out at pH 9.0; the amounts of DPCK and ATP were doubled to 9.8 mg and 30 mM, respectively; and the enzymes were removed using a 10K MWCO Amicon Ultra-15 filtration unit (EMD Millipore). An ATP regeneration system using 0.1 mg ml−1 pyrophosphatase (Roche), 30 mM phosphoenolpyruvate, and 0.1 mg ml−1 pyruvate kinase (Roche) was also included. The filtrate containing NH-CoA was purified on a preparative reverse-phase C18 HPLC (35 ml min−1; 0–4 min, 0% B; 4–9 min, 0–98% B, where A is 0.1% trifluoroacetic acid (TFA; Sigma-Aldrich) in H O and B is 0.1% in acetonitrile (ACN; Sigma-Aldrich)). NH-CoA was eluted at 7 min and lyophilized to dryness. Valine–amino-CoA (Val–NH-CoA)23 was synthesized by coupling 1 molar equivalent of NH-CoA with 8 molar equivalents of tert-butoxycarbonyl-l-valine-N-hydroxysuccinimide ester (Boc-Val-OSu; Sigma-Aldrich) in N,N-dimethylformamide (DMF; Sigma-Aldrich) with 4 molar equivalents of N,N-diisopropylethylamine (DIPEA; Sigma-Aldrich) overnight with stirring. Boc-Val-NH-CoA was purified using the above chromatographic profile and lyophilized to dryness, then deprotected using 1.5 ml 95% TFA, 2.5% H2O and 2.5% triisopropylsilane (TIPS; Sigma-Aldrich). The deprotection mix was agitated for 2 h at 25 °C in a thermomixer at 700 r.p.m. before being transferred to 20 ml ice-cold diethyl ether and incubated at −20 °C for 2 h. The solution was centrifuged and the pellet was redissolved in 5% aqueous ACN solution and purified with the same protocol as NH-CoA. Compound identity was verified by mass spectrometry and nuclear magnetic resonance (NMR) (Supplementary Data 1). Unmodified F–A–PCP was converted to Val–NH–F–A–PCP by incubating 25 μM apo-F–A–PCP with 5 μM of the promiscuous phosphopantetheinyl transferase Sfp, 0.25 mM Val–NH-CoA, 10 mM MgCl and 25 mM Tris (pH 7.0) for a minimum of 4 h at 25 °C. To remove Sfp for subsequent crystallization trials, the reaction mix was loaded onto a Superdex S75 10/300 GL (GE Healthcare Life Sciences) equilibrated in 25 mM Tris (pH 7.5), 150 mM NaCl and 2 mM βME. Inline size exclusion chromatography with small-angle X-ray diffraction (SEC–SAXS) data was collected on the G1 beamline at the Macromolecular Diffraction Facility at the Cornell High Energy Synchrotron Source32, 33 at 9.963 keV (1.244 A) at 7.89 × 1011 photons s−1. The X-ray beam was collimated to 250 × 250 μm and the sample cell path length was 2 mm. The G1 beamline was outfitted with a GE AKTA purifier with a GE Superdex 200 5/150 GL column and 50 μl sample loop. The column was equilibrated in 25 mM Tris (pH 7.5), 150 mM NaCl and 2 mM βME and the samples were centrifuged for 10 min before sample injection. Images were recorded on a Pilatus 100K-s detector and normalized using beam stop photodiode counts. F–A–PCP eluted in a single monomeric peak and eleven peak exposures were averaged using BioXTAS RAW software34. A buffer scattering curve was created by averaging the first eleven exposures after injection, and this scattering curve was subtracted from the F–A–PCP scattering curve to yield the corrected scattering curve for F–A–PCP. Ab initio models were generated by first creating pairwise distribution functions (P(r)) with GNOM35, leading to twenty independent bead models produced by DAMMIF36. Models were aligned, averaged, and filtered using DAMAVER37 assuming P1 symmetry. All DAMMIF models were included in the final DAMAVER model. They had a mean normalized spatial discrepancy (NSD) value of 0.82 ± 0.052. CRYSOL38 was used to check how well the final model fit with our crystal structures. Flexibility was analysed using EOM39, 40, whereby crystal structures of F, A , A and the PCP were used to generate a pool of 10,000 models. To obtain the crystal structures described in this study, genes from four species, of up to four domain constructs each (F, F–A, F–A(ΔA ) and F–A–PCP), were cloned and assayed for heterologous expression. Purification was performed for all highly expressing proteins and crystallization trials were performed, including trials using protein with affinity tags removed or retained, and in the presence or absence of a variety of ligands (ATP, AMPcPP, AMP, valine, THF, N5–f THF, phosphopantetheine, valine amino phosphopantetheine, valine vinyl sulfonamide adenylate, dead-end THF analogue). Up to 4,032 crystallization conditions were assayed per protein sample, and gave a total of ~50 ‘hits’, 6 of which were successfully optimized to allow structure determination. Together, 4 of these crystal structures (F–A in crystals of space group P4 2 2, F–A–PCP in R3:H, F–A–PCP–PPE–NH–Val in P2 and F–A–PCP–PPE in P3 2), plus an additional structure including ligands soaked into F–A P4 2 2 crystals, captured the states that represent every major step of the assembly-line synthesis in the LgrA initiation module and are presented here. The final crystallization conditions were optimized in 24-well sitting drop plates, with 2 μl protein sample plus 2 μl reservoir solution in the drop and a 500 μl reservoir volume, and are as follows. ‘F–A’ and ‘F–A soak’: protein LgrA F–A (10 mg ml−1) was crystallized using a precipitant solution of 2 M Na-formate, 0.1 M sodium acetate (pH 5.3) into space group P4 2 2. ‘F–A–PCP’ (open and closed states): protein LgrA F–A–PCP (5 mg ml−1) was crystallized using a precipitant solution of 0.92 M AmSO , 0.1 M bis-Tris (pH 5.5), 1% polyethylene glycol (PEG) 3350 into space group R3:H. ‘F–A–PCP–NH–Val’ (thiolation state): protein F–A–PCP–PPE–NH–Val (4.7 mg ml−1) was crystallized using a precipitant solution of 12% PEG 20,000, 0.1 M MES buffer (pH 6.7) into space group P2 . ‘F–A–PCP–PPE’ (formylation state): protein F–A–PCP–PPE (5.5 mg ml−1) was crystallized using a precipitant solution of 1 M AmSO , 0.1 M bis-Tris (pH 5.5), 3% PEG 3350 into space group P3 2. Solutions of mother liquor with increasing amounts of glycerol (5%, 10%, 25%) were used to replace the drop solution for cryoprotection. For soaking with the N5-f THF, valine and AMPcPP, 10 mM of each was included in the final cryoprotection solution and incubated for 30 min. (LgrA uses commercially available N5-f THF in addition to its natural substrate, N10-f THF6.) Crystals were flash-cooled in liquid nitrogen and diffraction data sets collected at 200 K using beamline 8 of the CMCF at the Canadian Light Source (λ = 0.979 Å) in Saskatoon, Canada. All data sets were integrated and scaled using the programs HKL-2000 (ref. 41) and iMosflm42. Structure determination of F–A in the P4 2 2 space group was performed by molecular replacement using a search model of the A domain from gramicidin Soviet synthetase17 (note that linear gramicidin and gramicidin Soviet are made by different NRPSs) with the A subdomain removed and side chains trimmed to the β-carbon, in the program Phaser43. Density for the F domain was visible in the resulting maps. Iterative building in the program COOT44 and refinement in the program Phenix45 produced the final F–A structure. This structure was then used as a search model to determine the structure of F–A–PCP in space groups P3 2, R3:H, and P2 by molecular replacement using the program Phaser, followed by iterative building in the program COOT and refinement in the programs Phenix and CNS46. The highest resolution shell CC* values are: P4 2 2, 0.845; P4 2 2 (soak), 0.897; P3 2, 0.883; R3:H, 0.822; and P2 , 0.822. The quoted resolution of each structure represents the half-data set correlation coefficient (CC 1/2) of the diffraction data42, 47. Multiple sequence alignments (MSAs) were constructed using Clustal Omega48 (http://www.ebi.ac.uk/Tools/msa/clustalo) and PROMALS3D49 (http://prodata.swmed.edu/promals3d), following database searches using BLAST50 (http://ncbi.nlm.nih.gov/blast). MSAs were drawn/edited using Jalview51 (http://www.jalview.org). PHYLIP (http://evolution.genetics.washington.edu) was used to make neighbour-joining trees bootstrapped with 100 replicates, and FigTree (http://tree.bio.ed.ac.uk) was used to draw them. WebLogo52 (http://weblogo.berkeley.edu) was used to draw sequence logos of residue groupings of interest. AmiGO53 (http://amigo.geneontology.org) was used to check for experimentally characterized proteins. Val–NH-CoA was verified by both mass spectrometry (calculated m/z [MH+]: 850.2304; measured m/z [MH+]: 850.2299) and 1H NMR [1H NMR (600 MHz, H O) δ 8.69 (d, J = 16.2 Hz, 1H), 8.46 (s, 1H), 6.25 (d, J = 5.9 Hz, 1H), 4.97–4.89 (m, 1H) 4.75–4.73 (s, 1H), 4.65–4.58 (m, 1H), 4.30–4.21 (m, 2H), 4.04 (s, 1H), 3.91–3.82 (d, J = 5.7 Hz, 1H), 3.75 (dd, J = 8.8 Hz, 1H), 3.68–3.61 (d, J = 5.9 Hz, 1H), 3.55–3.42 (m, 3H), 3.42–3.23 (m, 4H), 2.48 (m, 2H), 2.21–2.14 (m, 1H), 1.03–0.99 (m, 7H), 0.87 (s, 3H), 0.44 (s, 3H)].


A wearables company that has focused on producing sensors to assess and correct posture and optimize running performance has raised a $10 million Series B round, according to a company statement. The funding will be used to support its expansion with business to business partners in employer wellness, workplace safety and physical therapy. It will also support a technology platform for business to business partners. It is also partnering with companies with which it can integrate its sensor technology, such as sports apparel and running shoes. In a phone interview, Lumo Bodytech Co-Founder and CEO Monisha Perkash said the financing will be used to support product development among other areas. It plans to add at least eight staff across sales, marketing and product development. Until recently, the company has embraced a direct to consumer strategy, but it sees a lot of opportunity to diversify its business with corporate partners. Perkash said its Lumo Lift product is geared to improving posture, particularly to prevent back pain — the second most common reason people go to the doctor and the most common source of workplace injury claims for people under 45 years old. It has developed employer wellness partners that use Lumo Lift as well as standing desk manufacturer Workrite Ergonomics. It has also partnered with technology company Validic to aggregate data from devices so the data can be viewed in context with other activities. “Our sensors can be embedded in a variety of form factors  such as shirts, shorts, belts, and eyewear,” she said. It also comes with an app for iOS and Android networks. When users slouch, the sensor vibrates. WuXi Healthcare Ventures led the investment round, and other new investors included MAS Holdings, and SanDisk founder Eli Harari. The investment is part of a broader shift for WuXi, the corporate venture arm of China-based WuXi App Tech. It has previously focused mainly on life science investments. WuXi Healthcare also took part in 23andMe’s Series E round earlier this year. Perkash noted that its Lumo Run shorts with embedded sensors and apps for iOS and Android networks are available in pre-order. The sensors use algorithms to provide a detailed picture of the user’s running mechanics such as vertical oscillation, or bounce, cadence, and how much energy the user has expended. The feedback comes in realtime through a set earphones so users can correct themselves and track their performance on the app. “We are good at taking data and converting it into actionable insights,” Perkash said. The company launched its Lumo Lift product in China two weeks ago. Ge Li, CEO of WuXi AppTec said in a statement: “Wearables technology will fundamentally change the traditional healthcare industry and address a number of previously unmet wellness needs. People are looking for technology to support positive, proactive behavior change through advanced insights including real-time feedback and coaching.” Although wearables have generated interest in the mainstream market they have yet to gain traction on a massive scale. Dr. Ezekiel Emanuel is among those who have voiced pessimism that wearables companies will see implementation on a large scale for healthcare applications. Although wearables company partnerships with employer wellness businesses are not new, it will be interesting to see how effective Lumo Bodytech will be at converting workplace safety and physical therapy practices into customers.


Kickstarter alum, Lumo Bodytech announced on Tuesday that is has raised $10 million in Series B funding to launch its biomechanics platform. The Lumo Bodytech Biomechanics Wearables platform is a combination of hardware, software, and data that will enable companies to integrate human body movements into the next generation of wearables technology. The platform will be available to partners in the apparel, fitness, health, and workplace safety industries who seek to incorporate real-time motion tracking into their product offerings. “We’re now entering the next phase of the wearables industry, one where companies like Lumo harness lab-quality data and real-time coaching to solve meaningful health problems and support behavior change. With the support of our new investors including WuXi, a leader in research and development in healthcare applications, we can deliver a wearables platform that will ultimately help people address significant health, wellness and performance issues.” The announcement of Lumo’s new financing and the release of its platform come on the heels of the successful launch of the Lumo Run smart shorts and the reported ongoing success of its Lumo Lift posture sensors used by consumers and corporate wellness programs. The development of its previous products has helped Lumo Bodytech to develop technology and design innovations in the areas of connected hardware, software, advanced algorithms, and biomechanical applications. “Wearables technology will fundamentally change the traditional healthcare industry and address a number of previously unmet wellness needs. People are looking for technology to support positive, proactive behavior change through advanced insights including real-time feedback and coaching. Lumo is a platform for better health and living, whether that’s at home, in the office or at the gym.” WuXi Healthcare Ventures led the most recent round of financing with previous investors Madrona Venture Group, Innovation Endeavors, AME Cloud Ventures, and Innovalue Capital Ltd. MAS Holdings and SanDisk founder Eli Harari also join as strategic partners and new investors. “My patients have used the data from wearables and connected medical devices to help them manage their posture and activity level, which accelerated their recovery. Proper biomechanics, such as good posture, as well as moving regularly and avoiding extended sitting contributes to reductions in some of the most common conditions like neck, back and shoulder pain. If better posture and more regular movement was widely adopted it could lead to significantly decreased healthcare costs for companies and certainly a better quality of life for individuals.” Lumo has previously raised more than $6 million in equity financing and more than $2 million in crowdfunding.


News Article | November 3, 2015
Site: www.pehub.com

WuXi Healthcare Ventures led a $10 million Series B funding round in Lumo Bodytech, which makes wearable biomechanic products. Other investors in the funding round include Madrona Venture Group, Innovation Endeavors, AME Cloud Ventures, Innovalue Capital Ltd. MAS Holdings and SanDisk founder Eli Harari. Lumo previously raised more than $6 million in equity funding and more than $2 million in crowdfunding. Lumo Bodytech, the makers of the Lumo Lift posture coach and Lumo Run smart shorts, announced today that is has raised $10 million in Series B funding to launch the first wearables biomechanics platform. The Lumo Bodytech Biomechanics Wearables platform is a combination of hardware, software, and data that will enable companies to integrate human body movements into the next generation of wearables technology. The Lumo Platform will be available to partners in the apparel, fitness, health, and workplace safety industries who seek to incorporate real-time motion tracking into their product offerings. “We’re now entering the next phase of the wearables industry, one where companies like Lumo harness lab-quality data and real-time coaching to solve meaningful health problems and support behavior change,” said Lumo Bodytech CEO and co-founder Monisha Perkash. “With the support of our new investors including WuXi, a leader in research and development in healthcare applications, we can deliver a wearables platform that will ultimately help people address significant health, wellness and performance issues.” The announcement of Lumo’s new financing and the release of its platform come on the heels of the successful launch of the Lumo Run smart shorts and the ongoing commercial success of its Lumo Lift posture sensors used by consumers and corporate wellness programs. The successful development of its previous products has helped Lumo Bodytech to develop significant technology and design innovations in the areas of connected hardware, software, advanced algorithms, and biomechanical applications. Ge Li, CEO of WuXi AppTec states, “Wearables technology will fundamentally change the traditional healthcare industry and address a number of previously unmet wellness needs. People are looking for technology to support positive, proactive behavior change through advanced insights including real-time feedback and coaching. Lumo is a platform for better health and living, whether that’s at home, in the office or at the gym.” WuXi Healthcare Ventures led the most recent round of financing with previous investors Madrona Venture Group, Innovation Endeavors, AME Cloud Ventures, and Innovalue Capital Ltd. MAS Holdings, and SanDisk founder Eli Harari also join as strategic partners and new investors. William Updyke, DC chiropractor at the Cisco LifeConnections Health Center states, “My patients have used the data from wearables and connected medical devices to help them manage their posture and activity level, which accelerated their recovery. Proper biomechanics, such as good posture, as well as moving regularly and avoiding extended sitting contributes to reductions in some of the most common conditions like neck, back and shoulder pain. If better posture and more regular movement was widely adopted it could lead to significantly decreased healthcare costs for companies and certainly a better quality of life for individuals.” Lumo launched Lumo Lift in 2014 to great success, and since then Lumo Bodytech has collected the largest database of posture information in the world. Lumo has previously raised more than $6M in equity financing and more than $2M in crowdfunding. About Lumo Lumo Bodytech has developed a technology platform that leverages algorithms, software and smart sensors to optimize performance and address human biomechanics through the real-time tracking of body movement. Lumo’s products empower you to be the best version of yourself. Current Lumo Bodytech products include the Lumo Lift and Lumo Back posture coaches and activity trackers, as well as the Lumo Run smart running shorts. Lumo Bodytech is privately-held and headquartered in Palo Alto, CA. Information on Lumo Bodytech and their products is available at www.lumobodytech.com. To learn more about the Lumo Bodytech technology platform visit www.lumobodytech.com/platform. About WuXi Healthcare Ventures and WuXi AppTec WuXi Healthcare Ventures is a $250M independent venture capital firm affiliated with WuXi AppTec, which helps over 3000 leading life science companies to develop some of the most transformative therapies for the world. WuXi Healthcare Ventures invests globally in health companies and leverages the expertise and capabilities of WuXi AppTec to provide portfolio companies with resources needed to build, grow, and scale world-class companies.

Discover hidden collaborations