| Sign In to gain access to subscriptions and/or personal tools. |
Non-primate Lentiviral Vector Administration in the TMJ
1 Eastman Department of Dentistry, and Correspondence: * corresponding author, stephanos_kyrkanides{at}urmc.rochester.edu
Gene therapy is emerging as a novel treatment method for the management of temporomandibular joint disorders. The aim of this investigation was to study the effects of lentiviral vectors on the temporomandibular joint. Consequently, we injected into the articular joint space a defective feline immunodeficiency virus capable of infecting dividing as well as terminally differentiated cells with the reporter gene lacZ, the expression of which was studied by means of PCR, X-gal histochemistry, and β-galactosidase immunocytochemistry. Our results showed successful transduction of hard and soft tissues of the temporomandibular joint. Interestingly, a subset of primary sensory neurons of the ipsilateral trigeminal ganglion also stained positive for the reporter gene, presumably following uptake of the lentiviral vector by peripheral nerve fibers and retrograde transport to the nucleus. These findings suggest that lentiviral vectors can potentially serve as a platform for the transfer of anti-nociceptive genes for the management of temporomandibular joint pain.
Key Words: gene therapy • immunodeficiency virus • feline • beta-galactosidase • mouse • temporomandibular joint • trigeminal ganglion
Nociceptive innervation to the temporomandibular joint (TMJ) is primarily provided by the auriculotemporal nerve of the mandibular division of the trigeminal nerve (Sessle and Wu, 1991). A  and C nerve fibers, whose cell bodies are located in the posterolateral part of the trigeminal ganglion (Yoshino et al., 1998), project distally and terminate as non-encapsulated free nerve endings dispersed throughout the posterolateral part of the TMJ capsule (Bernick, 1962; Thilander, 1964; Frommer and Monroe, 1966; Klineberg, 1971), the posterior band of the meniscus, and the posterior attachment (Dreessen et al., 1990; Kido et al., 1991, 1993; Wink et al., 1992). In the quest for the development of new therapies for orofacial pain, gene therapy appears to be an emerging treatment method (Kuboki et al., 1999; Pohl and Braz, 2001; Baum et al., 2002). For example, it has been previously suggested that delivery of antisense oligonucleotides developed against nociceptive genes to appropriate tissues may offer alternatives in the design of novel treatments for pain management (Wu et al., 2001). We hypothesize that transfer of anti-nociceptive genes to sensory trigeminal neurons innervating the orofacial region may be achieved after injection of lentiviral vectors at the painful site, such as the TMJ, resulting in their uptake by free nerve endings and retrograde transport to the sensory cells’ nuclei. Previous studies demonstrated axonal retrograde transport of horseradish peroxidase from the TMJ to the central nervous system (Romfh et al., 1979; Capra, 1987), including the trigeminal ganglia (Yoshino et al., 1998). In evaluating lentiviral vectors as the basis for TMJ gene therapy, we used VSV-G pseudotyped feline immunodeficiency viral vectors (FIV) capable of stably transducing dividing, growth-arrested, as well as post-mitotic cells, since they are capable of transgene integration into the host’s genome (Poeschla et al., 1998). VSV-G pseudotyping of viral vectors confers a broad range of host specificity, including human and murine cells, since infection is mediated by the interaction of the viral envelope protein and a phospholipid component of the cell membrane, leading to membrane-fusion-mediated entry (Burns et al., 1993; Carneiro et al., 2002). Therefore, FIV vectors can potentially mediate sustained gene expression in non-dividing terminally differentiated trigeminal sensory neurons, a property unique to lentiviral vectors. The aim of the present study was to investigate the effects of viral-mediated gene transfer to neurons located in the trigeminal ganglion following local TMJ administration of a non-primate lentiviral vector.
Vector Construction and Packaging We used the defective, vesicular stomatitis (VSV-G) pseudotyped, feline immunodeficiency virus, FIV(lacZ), capable of transducing dividing, growth-arrested, as well as post-mitotic cells (neurons) with the reporter gene lacZ driven by the ubiquitous cytomegalovirus promoter CMV (Poeschla et al., 1998). The vectors were kindly donated to us by Dr. Wong-Staal, University of California at San Diego. A schematic description of the vector is depicted in Fig. 1A  . In addition, we constructed a control FIV( 'lac) vector carrying an inactive β-galactosidase by deleting the first 1000 bp of the lacZ gene (3.75 kb in total), including the transcription initiation site (Fig. 1A). Specifically, the FIV(lacZ) vector was digested in vitro with the SstII and Cla I restriction enzymes overnight at 37°C, followed by agarose gel purification. The ends of the backbone DNA were blunted with the T4 DNA polymerase (Invitrogen, Carlsbad, CA, USA) and ligated with T4 ligase (Invitrogen) according to manufacturer’s instructions. The FIV(lacZ) and FIV('lac) vectors were transiently co-transfected, along with the packaging and VSV-G vectors, into 293H cells (GIBCO/BRL) cultured in DMEM (Invitrogen) plus 10% FBS (Gemini, Woodland Hills, CA, USA), with the Lipofectamine 2000 reagent per manufacturer’s instructions (Invitrogen), and followed by a fresh medium change supplemented by non-essential amino acids (Invitrogen). Sixty hrs post-transfection, the supernatant was collected, filtered through a 0.45-mm Surfil®-MF filter (Corning Separations Division, Acton, MA, USA), aliquoted, and frozen until further use. Titering was performed on CrfK cells (American Tissue Culture Collection, Manassas, VA, USA) cultured in 24-well tissue culture plates, and assessed at at 5 x 107 blue forming units (bfu)/mL by X-gal histochemistry.
Animal Injections All methods pertinent to animal utilization were approved by the University Committee on Animal Resources. Specifically, 12 male mice, C57BL/6J, under surgical plane of anesthesia (ketamine 60 mg/Kg and xylazine 5 mg/Kg administered intraperitoneally), received a single injection of 5 x 106 FIV(lacZ) infectious particles (100 µL of stock solution) in the joint space of the right TMJ. Four additional mice received a single injection of 5 x 106 FIV( 'lac) infectious particles (100 µL of stock solution) in the joint space of the right TMJ. In brief, the hair of the skin covering the right TMJ was shaved, and the skin was cleaned with Betadine solution. An antero-posterior incision was made in the joint between the posterior end of the zygomatic arch and the ear cartilage, followed by a blunt dissection to expose the zygomatic arch and the posterior margin of the articular eminence. The joint space was not exposed during this procedure. The posterior margin of the eminence was identified by palpation, and a 1-mL tuberculin syringe with a 27 -gauge needle was used to inject the experimental solutions into the joint. This surgically assisted intra-articular injection technique was utilized to minimize leakage or spreading of the injectable solution beyond the articular space (Kyrkanides et al., 2002a). In addition, 2 mice that received 100 µL saline injection served as controls. Forty-five days following treatment, the mice were deeply anesthetized by pentobarbital (100 mg/Kg IP) and killed by transcardial perfusion of 4% paraformaldehyde in phosphate-buffered saline (PBS) (Kyrkanides et al., 2002a,b). The trigeminal ganglia and brain stem were dissected and sectioned at 20 µm by means of a freezing microtome. The TMJ joints were also dissected, decalcified in an EDTA-buffered solution, embedded in paraffin, and cut at 8-µm sections. All tissues were stored at –20°C until processed further.
X-Gal Histochemistry
Cell Counting
Immunocytochemistry Tissue sections from the temporomandibular joints were first deparaffinized by immersion in a series of xylines and alcohols, followed by antigen retrieval processing (95°C heating for 15 sec in 0.1 M Tris-HCl buffer, pH 8.9) and processing according to the aforementioned immunocytochemical method.
Polymerase Chain-reaction (PCR)
Intra-articular FIV Injection Resulted in Transduction of Hard and Soft Tissues FIV(lacZ) injection into the TMJ articular space resulted in transfer of the reporter gene lacZ via the lentiviral vector in cells located within the articular capsule. Specifically, cells of the TMJ meniscus, presumably fibroblasts, expressed bacterial β-galactosidase, as assessed by immunocytochemistry with appropriate polyclonal antibodies. In addition, cells located in the hypertrophic zone of the condyle—comprised primarily of cartilaginous cells as well as perivascular cells, including endothelial cells and possibly osteocytes—also stained positive for bacterial β-galactosidase (Fig. 2  ). There was a lack of β-galactosidase expression in the contralateral joints as well as the saline-injected animals. These results suggest that FIV successfully infected and stably transferred the reporter gene to cells of hard and soft TMJ tissues.
FIV Injection into the TMJ Resulted in Transduction of Trigeminal Neurons
Considerable Numbers of Trigeminal Neurons were Infected by FIV(lacZ) On average, the mouse ganglia were of the following dimensions: 1.5 x 2 x 3 mm. As described above, approximately 42 sections (20 µm thick) were produced from each ganglion and sequentially collected onto 3 glass slides, each containing representative ganglion sections 60 µm from each other. Consistently, we identified 4 sections containing X-gal (blue) cells on each glass slide, with an average of 93 (± 7.64 SD) blue cells per section. Therefore, we infer that there were approximately 93 cells x 4 sections x 3 glass slides = 1116 transduced neurons in each right-sided ganglion in the FIV-injected animals. No X-gal-positive cells were identified in the saline-injected animals. These results suggest that, from a total of 5 x 106 infectious particles injected into the articular TMJ space, approximately 103 nerve fibers were infected, resulting in lacZ expression, presumably following uptake of the lentiviral vector by peripheral nerve fibers and retrograde transport to the nucleus.
Pain alleviation from the orofacial region, including the TMJ, is based primarily on the use of pharmaceutical agents, often accompanied by many side-effects such as tolerance and physical dependence (Martin and Eades, 1961; Cox et al., 1968). Although gene therapy appears to carry promise in the development of pain therapy (Pohl and Braz, 2001), considerably more research is needed before gene transfer can be applied clinically as a strategy for chronic pain management (Baum et al., 2002). In a recent study, evidence was presented for the feasibility of direct gene delivery to the articular surfaces of the TMJ (Kuboki et al., 1999). Similarly, our results demonstrate that intra-articular injection of FIV(lacZ) resulted in successful gene transfer to articular TMJ surfaces as well as to the joint meniscus. Interestingly, VSV-G does not require interaction between the viral envelope protein and a specific membrane receptor, but instead interacts with a phospholipid component of the cell membrane, leading to membrane-fusion-mediated entry. This characteristic confers a broad host-cell range for VSV-G pseudotyped viruses (Burns et al., 1993; Carneiro et al., 2002). Therefore, it is possible that FIV vectors demonstrate higher infectivity for TMJ tissues than do previously described viral vectors (Kuboki et al., 1999), as well as resulting in prolonged transgene expression secondary to stable transgene integration (Poeschla et al., 1998). The efficacy of VSV-G pseudotyped FIV vectors to transduce peripheral tissues (Kang et al., 2002), as well as the brain (Blömer et al., 1997) and cerebellum (Alisky et al., 2002), has been previously demonstrated. However, limited information is available on the ability of non-primate lentiviruses to infect neurons retrogradely following peripheral administration. Our observations of cells staining positively for X-gal in the trigeminal ganglion ipsilateral to the site of injection suggest that FIV virions were taken up by peripheral nerve projections of trigeminal sensory neurons that lead to infection and expression of the reporter gene lacZ by these neurons. Novel strategies for the treatment of pain have been previously discussed (Wu et al., 2001), including the use of antisense oligonucleotides delivered to target cells that can bind to mRNAs encoding for nociceptive molecules. Therefore, VSV-G pseudotyped lentiviruses, such as the defective feline or human immunodeficiency virus, may serve as the platform for the transfer of anti-nociceptive genes to temporomandibular joint tissues as well as to the neurons that innervate these structures.
This work was supported in part by grants DE013860 and DE000471 from the National Institutes of Health, as well as by the American Association of Orthodontists Foundation. Received for publication March 18, 2003. Revision received August 15, 2003. Accepted for publication October 16, 2003.
Journal of Dental Research, Vol. 83, No. 1,
65-70 (2004)
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
|
|
 |
|
Sign In to gain access to subscriptions and/or personal tools.
TOP
ABSTRACT
INTRODUCTION
