Human degenerative disc disease (DDD) is characterized by progressive loss of human nucleus pulposus (HNP) cells and extracellular matrix, in which the massive deposition are secreted by HNP cells. Cell therapy to supplement HNP cells to degenerated discs has been thought to be a promising strategy to treat DDD. However, obtaining a large quality of fully functional HNP cells has been severely hampered by limited proliferation capacity of HNP cells in vitro. Previous studies have used lipofectamine or recombinant adeno-associated viral (rAAV) vectors to deliver human telomerase reverse transcriptase (hTERT) into ovine or HNP cells to prolong the activity of nucleus pulposus cells with limited success. Here we developed a lentiviral vector bearing both hTERT and a gene encoding green fluorescence protein (L-hTERT/EGFP). This vector efficiently mediated both hTERT and EGFP into freshly isolated HNP cells. The expressions of both transgenes in L-hTERT/EGFP transduced HNP cells were detected up to day 210 post viral infection, which was twice as long as rAAV vector did. Furthermore, we observed restored telomerase activity, maintained telomere length, delayed cell senescence, and increased cell proliferation rate in those L-hTERT/EGFP transduced HNP cells. Our study suggests that lentiviral vector might be a useful gene delivery vehicle for HNP cell therapy to treat DDD. © 2013 Orthopaedic Research Society. Published by Wiley Periodicals, Inc. J Orthop Res 32:159–166, 2014.
Lower back pain is a major public health problem in modern society. The most common cause for lower back pain is degeneration of intervertebral disc, which is often called degenerative disc disease (DDD). The DDD is characterized by progressive loss of human nucleus pulposus (HNP) cells and extracellular matrix. The HNP cells secrete aggrecan and collagen II, which are the massive deposition of the extracellular matrix during development. Therefore, cell therapy to supplement HNP cells to degenerated disc has been thought as a promising strategy to treat DDD. This strategy requires a large quantity of fully functional HNP cells.
HNP cells can be expanded in vitro with limited life span. After serial passages in vitro, HNP cells undergo replicative senescence marked with the changes on cell morphology, gene expression, and metabolism; and eventually they also lose their ability to synthesize and secret collagen II and aggrecan.[3, 4]
One of strategies to extend the expansion of functional HNP cells in vitro is to ectopically express human telomerase reverse transcriptase (hTERT) in HNP cells. The hTERT is a catalytic component of the telomerase. The telomerase maintains the length of telomeres of chromosomes by adding the telomere repeat sequence (TTAGGG). Ectopic expression of hTERT reconstitutes the telomerase activity, stabilizes telomere length, and extends the life span of various types of human cells,[5, 6] including HNP cells.
Previous studies have shown that delivering hTERT gene into HNP cells using lipofectamine and recombinant adeno-associated viral (rAAV) vector can extend ovine nucleus pulposus cells and HNP cells. The lipofectamine has some drawbacks such as transient expression of transgene and high cellular toxicity. Treating DDD, a chronic disease, needs long-term and continuous supplement of functional HNP cells. Thus, lipofectamine-mediated hTERT cannot meet its need. For these reasons, we previously developed a rAAV-hTERT vector to deliver hTERT to HNP cells. The rAAV-hTERT effectively mediates hTERT into HNP cells and extends the life-span of HNP in vitro up to 150 days. However, rAAV has a limited transgene packaging capacity (up to 5 kb). In addition, because this replication-deficient rAAV-hTERT exists in HNP cells as episomal concatamers,[10, 11] which cannot replicate during cell divisions, the expression of hTERT decreases gradually and cannot be detected at 120 days post viral infection.
Lentiviral vectors had been used to deliver genes into various types of cells, including both dividing and non-dividing cells. When infecting cells, lentiviral vectors incorporate themselves into the genome of transduced cells, allowing for stable transgene expression. In addition, lentiviral vectors have transgene packaging capacity of 8–9 kb. The self-inactivating lentiviral transfer vectors and the third generation packaging system which separates viral genes in three plasmids make lentiviral vector safer. In this study, we constructed a lentiviral vector bearing the hTERT coding sequence and DNA sequence coding for a green fluorescence protein under the control of a constitutive promoter EF1. This L-hTERT/EGFP vectors were used to mediate hTERT and EGFP into freshly isolated HNP cells and transduced HNP cells were analyzed for their proliferation and functions.
Construction of Lentiviral Transfer Vector and Viral Preparation
To construct a L-hTERT/EGFP lentiviral transfer vector, we inserted the hTERT cDNA (Geron Corporation, Menlo Park, CA) into a pPS-EF1-LCS-T2A-EGFP plasmid (Addgene, Cambridge, MA) at its ligase-free cloning site (LCS) according to the manufacturer's instructions. The VSV-G-pseudotyped lentiviral vectors were produced by cotransfecting 293T cells, and purified by ultracentrifugation as previously described.[14, 15] The titer (ifu/ml) of viral vector stocks was measured by a UltraRapid Lentiviral Titer Kit (Cell Biolabs, San Diego) according to the manufacturer's instructions.
Culture of Human Nucleus Pulposus Cells
Normal HNP tissue was obtained from a lumbar intervertebral disc of a 19-year-old man who was receiving the anterior fusion surgery for a burst fracture of L2 vertebrae. This procedure was carried out after an informed consent was assigned by the patient himself. After washing with PBS (HyClone, South Logan, UT)twice, the HNP tissue was cut into small pieces (about 1 mm2), then digested in DMEM medium (HyClone) with 0.025% collagenase II (Invitrogen, CA) for overnight at 37°C with 5% CO2. After 12 h, the HNP cells adherent to the flasks were harvested, plated into a 25 cm2 culture flask at a density of 105 cells/cm2, and grown in the DMEM containing 10% fetal bovine serum (Gibco, Grand Island, NY) and 1% penicillin/streptomycin at 37°C with 5% CO2. The growth medium was changed every 3 days.
In Vitro Viral Transduction
HNP cells grown in 6-well plates were allowed to reach 70–80% confluence before lentiviral transduction. After washing the cells with PBS, we added various amount of lentiviral vectors mixed with the growth medium in the presence of 6 µg/ml polybrene (Millipore, Darmstadt, Germany). Viral transduction was carried out in an incubator at 37°C with 5% CO2 for overnight. Seven days later, the transduction efficiency was measured by monitoring EGFP expression with flow cytometry(BD, Franklin Lakes, NJ).
Semi-Quantitative Reverse Transcription Polymerase Chain Reaction
To measure the expression of hTERT in lentiviral vector-transduced HNP cells, we used a semi-quantitative reverse transcription polymerase chain reaction (RT-PCR). Total RNA was extracted from HNP cells with a RNA extract kit (Qiagen, Hilden, Germany) according to the manufacturer's instructions. RNA samples were then reverse transcribed to cDNA using random dT primer and AMV reverse transcriptase (Promega, Beijing, China) at 42°C for 1 h. PCR was then carried out with the following cyclying conditions: 95°C for 5 min followed by 33 cycles of 94°C for 45 s, 56°C for 45 s, and 72°C for 45 s, and a final extension at 72°C for 10 min. The specific primer pairs for the hTERT were forward primer: 5′-GCC AGC ATC ATC AAA CCC-3′ and reverse primer: 5′-CCA CGA ACT GTC GCA TGT AC-3′). Another specific primer pairs for human glyceraldehyde-3-phosphate dehydrogenase (GAPDH) used as an internal control were forward primer: 5′-GAA GGT CGG AGT CAA CGG-3′ and reverse primer: 5′-GGA AGA TGG TGA TGG GAT T-3′). The PCR products were analyzed in 1.5% agarose gel stained with ethidium bromide (2 µg/ml) and visualized with a Densitograph System (Atto Iotechnologies, Inc., Tokyo, Japan). The hTERT value from each sample was obtained by normalizing it to the GAPDH value from the same sample using the CS Analyzer (version 2.01, Atto).
Western Blot Analysis
Cells were harvested and lysed in the lysis buffer (50 mmol/L Tris, pH 8.0, 1% NP-40, 100 µg/ml PMSF). Protein extracts (5 µg) were loaded in SDS-polyacrylamide gel and separated by electrophoresis. Then, proteins were transferred onto the polyvinylidene fluoride membrane (PVDF) (Millipore). After blocked with 5% non-fat dry milk in Tris-buffered saline (TBS), the membrane was incubated with primary mouse monoclonal antibodies against hTERT (Ab-3) (Abcam, MA) and human GAPDH (abcam) followed with secondary rabbit anti mouse antibodies conjugated with horseradish peroxidase (Chemicon, Temecula, CA).The proteins were visualized using Western blot chemiluminescent reagent according to the manufactur's instructions.
Telomerase Activity Assay
Telomerase activity was determined using the TRAPEZE XL Telomerase detection kit (Millipore Billerica, MA) according to the manufacturer's instructions. Briefly, protein extracts were prepared from polled cells from triplicate wells of 6-well plates (3 × 105 cells/well) and 1 µg of protein extract was used to measure the telomerase activity. Reactions were performed in a 96-well plate (Coring, Franklin St, NY) and each sample was measured with fluorometer with the excitation/emission parameters for fluorescence (485/535 nm) and sulforhodamine (585/620 nm). Relative telomerase activity was determined by the following calculation: [(ΔFL sample − ΔFL heat inactivated sample/ΔR)/(ΔFL/ΔR positive control)] × 100.
Telomere Length Assay
The cellular telomeric length was detected by TeloTAGGG Telomere Length Assay kit (Roche, Shanghai, China) according to the manufacturer's instructions. Briefly, genomic DNA was extracted from six pooled samples using the Genomic DNA purification kit (Gentra, Minneapolis, MN) according to the manufacturer's instructions. After digested with Rsa I and Hinf I, the prepared genomic DNA (2–3 µg) was separated on 0.6% agarose gels by electrophoresis (50 V for 4 h). The separated DNA was then transferred onto a nylon membrane (Roche, Frankfurt, Germany) for hybridization. The probed DNA was visualized by exposing membrane to a Hyperfilm. The mean telomere length was calculated using the formula: L = Σ (ODi)/Σ (ODi/Li), where ODi is the integrated signal intensity and Li is the DNA length.
Real-Time Polymerase Chain Reaction
We used real-time PCR to measure Type II collagens and aggrecan expression in HNP cells. Total RNA was extracted as described above from cells pooled from six wells with same conditions. The real-time PCR was performed in triplicates with a Rotor-Gene Thermal cycler (Corbett Research, Sydney, Australia) in Platinum Syber Green qPCR Super-Mix UDG (Invitrogen). The conditions for every PCR reaction were 95°C for 2 min followed by 50 cycles of 94°C for 30 s, 60°C for 30 s (annealing temperature was decreased 1°C per cycle for the first five cycles) and 73°C for 1 min. The sequence of primers used in real-time PCR were described before previous Wu's methods 6 (Wu et al., 2011).
Enzyme-Linked Immunosorbent Assay
To measure the protein levels of collagen 2 and aggrecan secreted by HNP cells in the growth medium, we used enzyme-linked immunosorbent assay (ELISA). The ELISA was carried out with an ELISA kit (BD Pharmingen, San Diego) according to the manufacturer's instructions. Triplicate growth media were collected for ELISA.
To determine senescent cells in cultured HNP cells, we used a Senescence Cell Staining Kit (CST, MA) according to the manufacturer's instructions. At 150 days after viral transduction, half million HNP cells were seeded in a well of six-well plate. Twenty-four hours later, cells were stained for -galactosidase activity at pH6.
Population Doubling Time
HNP cells of each passage were seeded onto the culture vessels by the density of 1 × 105 cells/ml. When reaching 85% confluence, cells were harvested, stained with trypan blue, and counted. Population doubling time (PDT) = (T–T0) × Log2/LogN–LogN0 where T–T0, the confluent time; N, the cell number at harvesting; N0, the cell number at seeding.
Metaphase spreads were prepared in 60-mm culture dish using standard techniques. Images of cells at metaphase were acquired by a video camera on a Leica microscope (MDE1850, Leica, Germany). The karyotypes of cells were analyzed by a computer-aided LEICA Q550CW image processing system (Leica, Germany).
Statistical analysis was performed using SPSS (version 3.0, IBM, CA). Experimental data were reported as the average number from at least three samples plus standard derivation. Comparisons among the data were performed using Student's t-test. Values of p < 0.05 were considered to be statistically significant.
The Efficancy of Lentiviral Vectors Transducing Human Nucleus Pulposus Cells
We first asked whether lentiviral vectors could efficiently transduce HNP cells. Freshly isolated HNP cells were infected with L-hTERT/EGFP at various MOIs. At day 7 post infection, percentages of HNP cells expressing EGFP increased along with MOIs, reaching up to 84.3% at MOI of 100. However, the percentage of cells expressing EGFP decreased along with time, dropping from 84.3% at day 7 to 1.37% at day 210 post infection (Fig. 1).
Expression of the hTERT Mediated by Lentiviral Vectors in Human Nucleus Pulposus Cells
We then asked whether the expression of the hTERT could be efficiently mediated by L-hTERT/EGFP in HNP cells. HNP cells infected with L-hTERT/EGFP at MOI of 100 were passaged in vitro continuously. At various time points post infection, we harvested cells to measure the mRNA levels of hTERT with reverse transcriptase PCR and the protein levels of hTERT with Western blotting. We found that the mRNA and protein levels of hTERT in HNP cells transduced with the L-hTERT/EGFP, but not with the L-EGFP, were detected at day 7 post viral infection. Unexpectedly, we found the both mRNA and protein levels of hTERT were gradually decreased along the time. At day 210 post viral infection, no mRNA, and protein of hTERT were detected (Fig. 2).
Extended Expansion Capacity of Human Nucleus Pulposus Cells Expressing hTRET
We then asked whether ectopic expression of hTERT restored the telomerase activity in cultured HNP cells. In L-EGFP- or mock-transduced HNP cells, the telomerase activity was low at any tested time points, even from the very beginning of culture (day 7 post viral infection). In contrast, in L-hTERT/EGFP-transduced HNP cells, the telomerase activities were dramatically increased although we observed a gradual decrease of the telomerase activities along the time. At day 210 post viral infection, the longest time point in our study, the telomerase activity in L-hTERT/EGFP-transduced HNP cells was still similar to the one in freshly isolated HNP cells (Fig. 3A).
We therefore asked whether ectopic expression of hTERT maintained the telomere length in cultured HNP cells. At day 7 post viral infection, no difference of the telomere length from L-hTERT/EGFP, L-EGFP, or mock-transduced HNP cells was observed. But, at day 120 and 210 post viral infection, the telomere length from L-hTERT/EGFP-transduced cells was apparently longer than the one from both L-EGFP- and mock-transduced cells although the telomere length from either group of cells was shorter at these time points than at day 7 post viral infection (Fig. 3B).
We further asked whether ectopic expression of hTERT delayed the senescence of HNP cells. Our experiments were terminated at day 214 post viral infection. At day 210 post viral infection, L-hTERT/EGFP, L-EGFP, or mock-transduced HNP cells were stained for β-galactosidase activity at pH6. We found that the percentage of senescent cells (blue) were significantly lower in L-hTERT/EGFP-transduced cells than in either L-EGFP- or mock-transduced cells (Fig. 3C and D).
Finally, we asked whether ectopic expression of hTERT altered the proliferation process of HNP cells. The population doubling time at each passage of L-hTERT/EGFP, L-EGFP, or mock-transduced HNP cells were measured. We found that from the passage 8, the population doubling time was significantly shorter in the L-hTERT/EGFP-transduced HNP cells that in either the L-EGFP- or the mock-transduced HNP cells (Fig. 4).
Taken together, these results suggest that ectopic expression of hTERT extented the expansion capacity of HNP cells in vitro.
Function of Human Nucleus Pulposus Cells Expressing hTERT
Normal functional HNP cells secret collage II and aggrecan. We asked whether HNP expressing hTERT were functional by measuring their mRNA using real-time PCR and protein levels using ELISA during the period of cell expansion in vitro. From day 7 to 150 post viral infection, the mRNA and the protein levels were significantly higher in the L-hTERT/EGFP-transduced cells than in the L-EGFP- or the mock-transduced cells (Tables 1 and 2).
|7 d||21 d||60 d||90 d||120 d||150 d||210 d|
|Collagen-II||8.54 ± 0.41a||7.12 ± 0.18a||5.82 ± 0.46a||4.27 ± 0.33a||3.02 ± 0.27a||1.71 ± 0.23a||0.99 ± 0.15|
|Aggrecan||2.75 ± 0.10a||2.43 ± 0.12a||2.09 ± 0.13a||1.76 ± 0.05a||1.38 ± 0.10a||1.21 ± 0.11||0.94 ± 0.05|
|7 d||3.92 ± 0.20a||2.59 ± 0.18||2.64 ± 0.17||87.02 ± 3.69a||30.96 ± 1.13||31.21 ± 0.93|
|21 d||3.03 ± 0.16a||1.83 ± 0.11||1.83 ± 0.12||79.05 ± 3.88a||15.73 ± 0.79||15.73 ± 0.79|
|60 d||2.41 ± 0.23a||1.44 ± 0.16||1.42 ± 0.18||69.97 ± 1.93a||10.94 ± 0.80||10.94 ± 0.80|
|90 d||2.01 ± 0.12a||1.23 ± 0.09||1.21 ± 0.11||53.82 ± 3.40a||9.62 ± 0.42||9.62 ± 0.42|
|120 d||1.47 ± 0.12a||1.06 ± 0.07||1.08 ± 0.14||37.45 ± 2.48a||8.92 ± 0.52||8.92 ± 0.52|
|150 d||1.10 ± 0.09||0.97 ± 0.07||1.00 ± 0.05||19.33 ± 2.96a||7.27 ± 0.58||7.27 ± 0.58|
|210 d||0.90 ± 0.09||0.87 ± 0.06||0.93 ± 0.06||7.26 ± 0.54||6.81 ± 0.43||6.81 ± 0.43|
When HNP cells transduced with L-hTERT/EGFP were expanded to day 210 post viral infection, some cells were used to analyze their karyotypes. In the 75 analyzed cells, 53.3% (40/75) of cells had normal karyotypes (46, XY), 46.7% (35/75) of cells had aneuploidy of 45, XY (32 cells), or 44, XY (3 cells). In 35 out of 75 cells, chromosomal breakage was observed. No polyploidy karyotype was detected (Fig. 5).
Our previous study has shown that rAAV-mediated ectopic expression of hTERT in HNP cells prolongs their activity in vitro up to 120 days. In this study, we used the third generation lentiviral vectors to deliver hTERT to freshly isolated HNP cells and found that lentiviral vector-mediated ectopic expression of hTERT in HNP cells extended the activity of HNP cells in vitro up to 210 days, almost being twice as long as rAAV did. The lentiviral vector developed in this study bore both hTERT and EGFP under the control of the promoter derived from elongation factor 1 gene. EGFP expression in HNP cells allows us to monitor transgene expression easily. The following evidences supporting our conclusion were generated using this L-hTERT/EGFP. First, we demonstrated that L-hTERT/EGFP efficiently mediated transgens into HNP cells. At 100 of MOI, 84.3% of freshly isolated HNP cells expressed EGFP and at day 210 post viral infection, 1.4% cells still expressed EGFP (Fig. 1). Consistent with this observation, mRNA and protein levels of hTERT in these cells were also detected up to day 150 post viral infection (Fig. 2). Second, we showed that in those L-hTERT/EGFP transduced HNP cells, the telomerase activity was restored, the telomere length was maintained, and the cell senescence was delayed (Fig. 3) Third, the L-hTERT/EGFP transduced HNP cells had shorter population doubling time than L-EGFP or mock transduced HNP cells from passage 8. Fourth, the L-hTERT/EGFP transduced HNP cells secreted much more collagen-II and aggrecan than L-EGFP or mock transduced HNP cells. Most importantly, the improved functions of L-hTERT/EGFP transduced HNP cells were observed even after 150 days post viral infection.
Although in our study lentiviral vector-mediated ectopic expression of hTERT extended the activity of HNP cells in vitro up to 210 days, almost twice as long as rAAV did, we unexpectedly observed that the EGFP or hTERT expression in these cells decreased along the time. Consistent with this observation, the activity of these cells, including the increased telomerase activity, extended telomere length, decreased cell senescence, and their ability to secrete collagen II and aggrecan were also found to be lost gradually along the time. Lentiviral vectors have an ability to integrate into the genome of target cells, achieving stable transgene expression. The results obtained from our study appeared to not be what we expected. Several possible explanations account for our observations. First, lentiviral vectors might be existed in their target cells as episomes, which cannot replicate themselves and will be lost during cell division. Second, the toxicity of EGFP has been reported in some types of cells.[17, 18]
Because a strong promoter, EF1, was used to drive the transgenes in our vector design, high levels of EGFP might cause cells to expel integrated transgene expression cassette, or to silence the EF1 promoter, or to cause cell death. Third, EF1 promoter used in our vector might be silenced cellular mechanisms. Promoter silencing in lentiviral transduced cells were often reported.
Studies need to be done to understand the underlying mechanisms by which the expressions of transgenes in L-hTERT/EGFP transduced HNP cells were not stable in order to improve the vector's performance.
We found that about half of analyzed cells had abnormal karyotypes. Most of them were aneuploids with one or two missing chromosomes. The high percentage of aneupliod karyotypes was also observed in those HNP cells of passage 2, which were not infected with lentiviral vectors (data not shown). In our previous study in which rAAV-transduced HNP cells with similar percentage of aneuploids were subcutaneously injected into athymic nude mice, we did not observe tumor formation, suggesting these cells may not be tumorigenic cells.
Safety is the major concern when lentiviral vectors are used to deliver the hTERT gene in our approach. The safety concern of our approach is from two aspects. First, the hTERT was over-expressed in HNP cells. High expression of the TERT is usually detected in highly proliferative cells, such as stem cells, tumor cells, etc., but not in normal somatic cells.[20, 21] Thus, cells with high telomerase activity are often thought to be tumorigenic. However, high telomerase activity in cancer cells may be the secondary event. For example, ovarian cancer cells have short length of telomeres and high telomerase activity, suggesting that these cancer cells have gone through many cell divisions and telomere shortening processes before high telomerase activity occurs.[22, 23] In fact, the over-expression of TERT gene has been extensively used to generate immortalized cells without tumorigenecity.[24, 25] Second, a lentiviral vector is derived from HIV-1, which causes acquired immunodeficiency syndrome (AIDS) in human. However, we used the newly developed third generation of lentiviral viral vectors which have decreased genotoxicity, relatively to the retroviral vectors.[26, 27] In this viral vector system, the transfer vector does not bear any viral gene; and three essential viral packaging genes, Gag, Pol, and Rev, as well as a heterologous envelope gene, VSV-G, are placed in the three packaging plasmids (pGag/Pol, pRev, pVSV-G). These packaging plasmids do not share any homologous region with the transfer vector, thus greatly reducing the risk of the formation of replication competent lentivirus (RCL). In addition, the U3 in 5′ LTR of viral transfer vector is replaced with a RSV promoter to prevent the Tat-dependent viral replication; the U3 in 3′ LTR is deleted to block the self-transcription of the viral transfer vector [a so-called self-inactivated (SIN) design]. These modifications greatly enhance the safety of using lentiviral vectors for gene therapy. Finally, although lentiviral vectors preferentially integrate into transcription active regions in a host genome, their integration sites are randomly localized in the entire gene. This feature also makes it superior to a retroviral vectors because it preferentially inserts at the regulatory regions of transcription active genes, potentially activates proto oncogenes, and increases the risk of generating tumors.
In conclusion, we demonstrated that lentiviral vectors can be used to efficiently deliver hTERT gene into HNP cells and to extend these cells expansion in vitro without the loss of their function. Safety is one of the most important problems in the gene therapy, although our study didn't show a satisfy results in the security for clinical use, we believe that lentivirus vector is a useful tool for basic research in DDD.
The research leading to these results has received funding from National Natural Science Foundation of China (Grant No. 81171740). The authors would like to thank Minghong Jiang and Jun Yan for their expertise in helping accomplish the PCR and western-blot test.