View File - Neurotech USA

229-C0084
PROTEOMIC MASS SPECTROSCOPY PROFILING OF PROTEINS SECRETED
FROM AN ENCAPSULATED CELL TECHNOLOGY (ECT) IMPLANT DEVICE
L. Orecchio , M. Rivera , E. Brissette , C. McGovern , A. Nystuen , K. Kauper
1
1
1
1
1
1
Neurotech Pharmaceuticals Inc., Cumberland, RI1
1. PURPOSE
3. RESULTS
3. RESULTS
Non-clinical and clinical studies have demonstrated the preliminary safety
and efficacy of a novel implant device (NT-503-3) (Figure 1)
encapsulating a retinal pigment epithelial cell line (NTC-203-910)
engineered to continuously produce
a homodimer, Fc chimera, VEGF
inhibitor (910-VEGFR). Analytical
techniques, including proteomic
mass spectroscopy (MS), were used
to characterize the 910-VEGFR
protein and identify the most
abundant proteins naturally secreted
from the encapsulated NTC-203-910
cell line.
Figure 1. NT-503-3 implant device
2. METHODS AND MATERIALS
5 devices from 3 consistency lots of NT503-3 product, one week post
encapsulation were incubated in culture media for 24 hours. Device
conditioned media (CM) was analyzed by VEGF direct binding ELISA,
SDS PAGE gels and western blots and MS.
For MS analysis CM samples were buffer exchange with a 0.5 ml, 3 kDa
Amicon filter (Millipore). Nano-LS/MS/MS using a trapping configuration
with a 4 hour reverse phase gradient identified the most abundant
proteins secreted by the encapsulated cell line. Relative abundance of
each protein (NSAF) was determined by normalizing raw MS spectral
counts (SpC) for molecular weight and calculating relative percentage of
total spectral counts (NSAF). Standard statistical methods, mean,
standard deviation (SD), percent coefficient of variation (%CV), and
coefficient of determination (R2) compared the NSAF profile of secreted
proteins from the 15 NT-503-3 devices.
Figure 5. Clinical lot protein profile linear regression line and R2. The Graph axis is expanded to
allow visualization of the majority of proteins. The graph is missing the top two abundant proteins.
Figure 2. 4-12% Bis-Tris SDS PAGE gels of non-reduced NT503-3 device CM. 40 ng of 910-VEGFR loaded per well.
A. SYPRO Ruby total protein stain. B. Western with goat anti-IgG1-Fc-HRP (Sigma A0170). C. Western with goat
anti-VEGFR1 primary (R&D anti-Flt1 AF321) and bovine anti-goat IgG-HRP secondary (Jackson Immuno 805-035-180).
Preliminary gene ontology analysis of the MS detected proteins secreted from 15 devices
from 3 lots of NT-503-3 product appears to be consistent with normal retinal pigment
epithelial cell line metabolic, development and organizational function (Figure 6).
MS total SpC counts were compared within and between lots. The intralot and interlot variation is <14% (Figure 3). 352 proteins were reliably identified technically with the 4 hour gradient mass spec method employed
(SpC>5). These proteins accounted for over 90% of the proteins. NSAF values indicating relative abundance
for the 352 proteins were calculated showing 910-VEGFR is the most abundant protein secreted. Pair-wise
linear regression analysis of the protein profile (NSAF) resulted in R2 values >93% indicating stable protein
expression among devices and lots of NT503-3 product (Figure 4).
3. RESULTS
VEGF direct binding ELISA determination of 910-VEGFR output was
within the specification range of manufacturing product stability. The
mean 910-VEGFR released per lot (n=5) was 10,831+610, 9,355+1,445,
11,101+1508 ng/device/day.
Non-reduced CM samples were electrophoresed on SDS PAGE gels. A
prominent, visible band at 132-147 kDa was observed. Western blots
using anti-hVEGFR and anti-hIgG1-Fc antibodies identified this band as
intact 910-VEGFR with no appreciable breakdown. The only other
prominent band was from the media, as shown by CM of a device with no
cells. Only a few other faint, discrete bands are discernible on the Sypro
Ruby stained gel (Figure 2).
The authors would like to acknowledge MS Bioworks, based in Ann Arbor, Michigan
for excellent technical services.
Figure 6. Gene Ontology Function Chart
4. CONCLUSIONS
Figure 4. Comparison MS protein profile.
Figure 3. Variation total SpC values from a 4hr MS
gradient among 3 lots.
For SpC
comparison, all the detected proteins were compared.
For NSAF comparison, the 352 most reliably detected
technically proteins were compared.
Six clinical lots (n=3 devices/lot) were examined by the same MS technique. The protein profiles (NSAF of
352 technically reliably identified proteins) were compared to the consistency lots using the same pair-wise
linear regression analysis. The R2 for the secreted proteins was greater than 0.9 suggesting a consistency
in the quality and quantity of the most abundantly secreted proteins from NT503-3 devices (Figure 5).
A single, intraocular ECT implant delivering 910-VEGFR was designed to provide
comparable anti-VEGF therapy to standard-of-care treatments while eliminating the
burden of frequent injections in patients with neovascular AMD. MS proteome profiling
identified, in addition to 910-VEGFR, the most abundant proteins stably secreted by
NT-503-3 devices as lower-level, naturally occurring proteins normal to human retinal
pigment epithelial function.
Lisa Orecchio
[email protected]