Identification and characterization of human nucleus pulposus cell specific serotypes of adeno-associated virus for gene therapeutic approaches of intervertebral disc disorders

Ethics statement and NP tissue recruitment

NP cell isolation and monolayer cell culture

NP tissues were washed in PBS and separated from anulus fibrosus (AF) on the basis of their macroscopic morphology (identification of the innermost lamellar rings of the AF). NP tissues were finely minced into small fragments of approximately 2 mm³ and digested with 0.02 % w/v pronase (Sigma-Aldrich) (37 °C, 5 % CO₂, 1 h) in 20 ml DMEM (Dulbecco's Modified Eagle's Medium), containing 1 % penicillin/streptomycin, 1 % glucose and 10 % FCS (fetal calf serum) (Sigma-Aldrich). After filtration through sterile 75 gm nylon mesh filters (Sigma-Aldrich) and centrifugation of the supernatants (1000 x g, 2 min), pellets were resuspended in 20 ml DMEM and digested with 0.02 % w/v collagenase II (Sigma-Aldrich) and 100U hyaluronidase (Sigma-Aldrich) (37 °C, 5 % CO₂, 3 h). Following filteration and centrifugation of supernatants, pellets were resuspended in 10 ml DMEM and cultured in 25 cm² tissue culture flask (Sigma-Aldrich) (37 °C, 5 % CO₂, 2 weeks) by changing the medium every two days. Depending on the amount of cells that could be isolated from the tissues, NP cells could reach 100 % confluentbetween two and three weeks. Cells were cryopreserved at −196 °C in DMEM containing 30 % FCS and 15 % dimethyl sulfoxide (DMSO) (Sigma-Aldrich).

Helper virus free production of scAAV serotypes

The AAV helper virus free system provides a safer and more convenient gene delivery system. Most of the adenovirus gene products (E2A, E4, and VA RNA genes) required for the production of infective recombinant AAV (rAAV) particles are cloned on the helper plasmid of the respective serotype (pDP1rs, pDP2rs, pDP3rs, pDP4rs, pDP5rs, pDP6rs, and pDP8rs) (PlasmidFactory, Bielefeld, Germany). For vector production of scAAV serotypes, HEK293 cells were cultured in DMEM and passaged 2 times prior to transfection. 5 × 10⁶ cells were grown in 15 cm culture dish of 20 ml culture medium to a confluence of 70 - 80 %. 30 μg of the scAAV shuttle plasmid [40], encoding the expression cassettes of GFP under the control of human CMV promoter, and 96 μg helper plasmids of the respective serotype were used for co-transfection. 30 μg of the scAAV shuttle plasmid and 96 μg of the respective helper plasmid were added in 2.5 ml of 300 mM calcium phosphate (Sigma-Aldrich), gently mixed with 2.5 ml of 2 x HBS (Hepes Buffered Saline) (Sigma-Aldrich) and directly pipetted to the culture dish. Following incubation (37 °C, 5 % CO₂, 6 h) the transfection medium was changed with prewarmed DMEM containing 2 % FCS. Transfected cells and culture medium were harvested 72 h after transfection and centrifuged (2000 x g, 5 min). Pellet was resuspended in 2.5 ml of serum-free DMEM, treated with benzonase (250 U/ml, 37 °C, 1 h) (Merck Millipore) and subjected to four rounds of freeze/thaw cycles by alternating the tube between the dry ice-ethanol bath and the 37 °C water bath. AAV supernatant was then collected by centrifugation (8000 x g, 30 min) and stored at −80 °C for subsequent purification.

Purification and quantification of scAAV serotype vectors

Purification of scAAV serotype vectors was performed as previously described [41]. Briefly scAAV vectors were purified from benzonase treated and cleared freeze/thaw-supernatants by iodixanol (Sigma-Aldrich) gradient centrifugation. Iodixanol was then removed by running the iodixanol fractions through PD10 gel filtration columns (GE Healthcare). The eluate was collected in 10 fractions of 1 ml and fractions 4 to 6 were pooled and used for quantification.

Quantitative PCR was applied for quantification of scAAV vector titers using LightCycler 480 (Roche Applied Science) and the TaqMan Gene Expression Master Mix (Life Technologies). PCR reactions were performed in 20 μl of final volume using 1× master mix, supplemented with 100 nM sense, 100 nM antisense primers of GFP and 2 μl of the standard template DNA. The primers GFP-sense:ACGGCGACGTAAACGGCCAC and GPF-antisense:GCGAAGCACTGCACGCCGTA were used. Standard sample and negative control were run in three replicates of a 96 well-plate. Linearized scAAV vector (shuttle plasmid cleaved with KasI) and genomic DNA of scAAV vector (scAAVstd) were used as standard as previously described [42]. The PCR programm had an initial denaturation step at 95 °C for 10 min, 40 cycles of denaturation at 95 °C for 15 s, an extension at 60 °C for 1 min and a melt curve stage (65 °C to 95 °C, increment 0.5 °C). Data analysis was performed using the Applied Biosystems StepOne software v2.1 (Life Technologies).

Transduction of human NP cells with scAAV serotype vectors

NP cells from 12 lumbar discs of degeneration grade III-V (Table 1) were seeded in 24-well plates at a density of 1 × 10⁵ cells per well (about 50 % confluent) in a volume of 500 μl culture medium containing 1 % FBS. The next day NP cells of each lumbar disc were initially transduced with scAAV1-8 at different doses of vector genome copy per seeded cell (5 vg/c, 50 vg/c, 500 vg/c and 5000 vg/c). After preliminary screening additional transductions were performed with selected serotypes and effective dose to quantify the transduction efficiencies, evaluate potential cell toxicities and determine prolonged transgene expressions.

Evaluation of transduction efficiencies using fluorescence microscopy and FACS

For preliminary screening of scAAV serotypes having human NP tissue tropism, transduced NP cells were visualized using an AxioVert. A microscope (Carl Zeiss). Fluorescence micrographs were acquired every 2 days for the first 24 days and weekly up to 48 days of post-transduction. Selected scAAV serotypes and viral dose, which efficiently transduced NP cells, were used for further quantification of transduction efficiencies by FACS. For FACS analysis NP cells were seeded as described above and transduced with selected scAAV serotypes and viral dose. Transduced cells were harvested on day 8, 16, 24, 32 and 48, and 1 × 10⁵ cells per sample were counted. The proportion of GFP-positive cells was determined using MoFlo cell sorter (Beckman Coulter) according to the manufacturer’s protocol. Briefly, a MoFlo cell sorter with a 100-mm flow cell tip and a flow rate of 12000 events per second, with an extention wavelength of 488 nm and a laser power of 110 W was used.

Cell viability assay

To analyse the cytotoxic effects of scAAV serotypes, which are able to effectively transduce human NP cells, the number of viable NP cells were quantified using 3-(4, 5-dimethylthiazolyl-2)-2,5-diphenyltetrazolium bromide (MTT Assay Kit, Molecular Probes). Transduced and non-transduced (negative control) NP cells were harvested after 8, 16, 24, 32 and 48 days of post-transduction. NP cells were trypsinated, centrifugated and pellets were washed twice with 1 x PBS and resuspended in 250 μl culture medium. Duplicates of 100 μl suspended cells were plated into flat-bottomed 96-well plate. Duplicate control wells of medium alone were added to provide the blanks for absorbance readings. After incubation for recovering (37 °C, 5 % CO₂, 24 h), 10 μl MTT reagent was added to each well and incubated again for 3 h. Then 100 μl of the SDS-HCl (Sodium dodecyl sulfate - hydrochloric acid) solution was added to each well for further incubation of 4 h. The absorbance was measured at 570 nm in a microtiter plate reader Infinite 200 (TECAN) and the average value of the blank duplicate readings was subtracted from the average values of the sample duplicate readings. Number of viable NP cells was calculated from the standard curve.

Three-dimensional NP cell culture

For three-dimensional culture (3D) of NP cells the collagen type I scaffold of 24-well plate format was used (Viscofan Bioengineering). 250 μl PBS (pH 7.3 without Ca2⁺/Mg2⁺) was added to each well for attaching of the scaffold to the bottom of the well. The scaffold was taken out of the blister with sterilized forceps, placed onto the PBS without submerging and incubated at room temperature for 20 min. After removing of the remaining PBS, the plate was left in the operating laminar flow hood overnight. The scaffold was equilibrated with 250 ml pre-warmed culture medium (37 °C, 5 % CO₂, 10 min) and monolayer scAAV transduced NP cells were transferred to collagen I scaffold on day 8 of posttransduction. NP cells were cultured in 500 μl culture medium (37 °C, 5 % CO2) by changing the culture medium every two days. As required, NP cells were harvested by digestion of the scaffold with 0.02 % w/v collagenase II in 250 μl medium (37 °C, 5 % CO₂, 1 h). Cell suspension was filtered through sterile 75 gm nylon mesh filter and supernatant was centrifuged (1000 x g, 2 min). Then pellet was washed twice in 1 ml PBS (1000 x g, 2 min) and processed for assays of inflammatory, catabolic and matrix proteins expressions.

Enzyme-linked immunosorbant assay of inflammatory, catabolic and matrix proteins

To examine the impact of scAAV vectors on the expression levels of inflammatory, catabolic and matrix proteins in NP cells, ELISA was performed on 100 μg of total protein extracts from each sample for each independent experiment. As described above, NP cells were seeded and transduced with the favoured scAAV serotype at its effective dose. On day 8 of post-transduction NP cells were harvested and cell pellet was washed with 1 x PBS, resuspended in 500 μl culture medium and three-dimensional cultured for 30 days. Cells were harvested by digestion of the scaffold as described above and for protein isolation pellet was washed twice in cold 1 x PBS (2500 x g, 5 min) and resuspended with 100 μl cold radio-immunoprecipitation assay (RIPA) buffer (Sigma-Aldrich) containing protease and phosphatase inhibitor cocktails (Sigma-Aldrich). After sonication at 50 % pulse for 30 s, the mixture was shacked gently on ice (15 min), centrifuged (14000 x g, 4 °C and 15 min) and supernatant was transferred to new tube for protein quantification. Protein concentration in sample was determined using BCA Protein Assay Kit (Thermo Scientific) according to the instruction manual. For measuring the concentration of inflammatory, catabolic and matrix proteins in NP cells ELISA kits were used (R&D Systems, Uscn Life Science Inc). Assays were performed according to the instruction manuals. The analysed target proteins include: the inflammatory cytokines IL-1β (interleukin-1β) and TNF-α (tumornekrosefaktor-α); the catabolic factors ADAMTS-4, ADAMTS-5 (a disintegrin and metalloproteinase with thrombospondin motifs) as well as matrix proteins aggrecan and collagen type II.

Statistical data analysis

To estimate the reliability of the MRI evaluations of IVD degeneration, Landis and Koch based interpretations with κ statistics and agreement percentage among two observers (interobserver reliability) were used [38, 39]. The software IBM SPSS Statistics 20, Armonk New York USA was used for statistical analysis. 1-way ANOVA and pairwise comparisons were used to compare scAAV treated and non-treated cells. Significance was set at P < 0.05.