BACKGROUND

High tibial Osteotomy (HTO) realigns the forces in the knee to slow the progression of osteoarthritis. This study relates the changes in knee joint biomechanics during level gait to glutamate signalling in the subchondral bone of patients pre and post HTO. Glutamate transmits mechanical signals in bone and activates glutamate receptors to influence inflammation, degeneration and nociception in arthritic joints. Thus glutamate signalling is a mechanism whereby mechanical load can directly modulate joint pathology and pain.

Glutamate is a neurotransmitter that transmits mechanical signals in bone (1) and activates glutamate receptors and transporters, in bone, cartilage, meniscus and synovium (2). Glutamate receptor activation influences inflammatory, degenerative and nociceptive pathways in arthritic joints (2). Thus glutamate signalling is a mechanism whereby mechanical load can directly influence joint pathology and pain. We have investigated components of glutamate signalling in the subchondral bone of patients with osteoarthritis to determine which are expressed and whether this varies in anatomical regions subject to different loads. Subchondral bone was sampled from tibial cuts derived from total knee arthroplasty (n=2, TKR, Kellgren Lawrence grade 3) and from tibial drill hole sites from high tibial osteotomy (n=5, HTO, KL grades 2 and 3) for osteoarthritis. RNA was extracted, reverse transcribed and RT-PCR performed for a housekeeping gene GAPDH, a glutamate transporters (EAAT-1, EAAT1ex9skip), glutamate receptors (NR2A and KA1), a bone matrix protein, osteocalcin, and signaling molecules (osteoprotegerin [OPG], RANKL). We found differential mRNA expression in different regions of subchondral bone. In one TKR patient, EAAT-1 expression was significantly reduced in the anterior zone versus the middle or posterior zones of the tibial plateau (ANOVA, p<0.001). HTO bone cores were subdivided medial/lateral and anterior/posterior. Good quality RNA was obtained from bone cores removed from drill holes during HTO surgery, with GAPDH, osteocalcin, EAAT-1, EAAT1ex9skip, NR2A, KA1, OPG and RANKL mRNA expression detected. In one patient, comparison of gene expression in bone cores obtained pre and post HTO revealed that EAAT1ex9skip was rarely detected in post-op bone whereas KA1 was rare in pre-op bone. This differential mRNA expression may be due to the altered loading through the joint caused by the osteotomy, although these on/off differences need to be quantified to confirm this.

We have shown that glutamate transporters and receptors are expressed in human subchondral bone. Activation of these receptors and transporters by the increased synovial fluid concentrations of glutamate released in arthritis will influence pathological changes and nociception. In some patients, glutamate transporter mRNA expression appears to vary with anatomical location in bone, or after HTO surgery, consistent with our original discovery of this transporter as mechanically-regulated in bone (1). If glutamatergic signaling is mechanically regulated in the human knee, this will vary during arthritic disease progression and after joint realignment, providing a direct mechanism linking mechanical loading through the joint to pathology and pain in arthritis.

Mechanical loading is a potent stimulator of bone formation. A screen for genes associated with mechanically-induced osteogenesis implicated the glutamate transporter GLAST-1 (1), in the mechanoresponse. We are investigating whether modulation of glutamate transporters represents a potential anabolic therapy in bone. Bone cells express functional components from each stage of the glutamate signalling pathway and activation of ionotropic glutamate receptors on osteoblasts can increase bone forming activity (2). Five high affinity Na+-dependant excitatory amino acid transporters (EAATs 1-5) regulate glutamatergic signalling. EAAT1 (GLAST-1) is expressed by osteocytes and bone-forming osteoblasts in vivo.

We quantified transcripts for EAATs 1-3 and two splice variants (EAAT1a and EAAT1ex9skip) in human osteoblasts (MG63, SaOS-2 and primary) using real time-PCR. EAAT1a expression was very low whilst levels of the dominant negative EAAT1ex9skip were much higher in all cell types. EAAT1 and EAAT3 proteins were detected by immunofluorescence. We also demonstrated that glutamate transporters function in human osteoblasts. Sodium-dependent 14C-labelled glutamate uptake, sensitive to pharmacological EAAT inhibitors (t-PDC, TBOA) and extracellular glutamate concentration (10-500μM) was detected in MG63 and SaOS-2 cells.

To determine whether modulation of EAATs can influence bone formation, we used pharmacological inhibitors of EAATs 1-5 (t-PDC and TBOA) and also over-expressed EAAT1exon9skip using antisense oligonucleotides (AONs) targeted to splice donor sequence of exon 9. Experiments were performed in 0-500μM glutamate. Pharmacological inhibition of EAATs over 5-21 days increased alkaline phosphatase activity and mineralisation of SaOS-2 cells and human primary osteoblasts. Over-expression of EAAT1ex9skip significantly increased cell number and decreased cell death as well as significantly increasing PCNA, Osteonectin and Type I collagen mRNAs in MG63 cells. Furthermore, over-expression of EAAT1ex9skip increased mean alkaline phosphatase activity over 48hrs in SaOS-2 cells.

These data show that EAATs are expressed and functional in osteoblasts and that pharmaceutical and genetic inhibition of their activity increases bone formation. These mechanically regulated glutamate transporters are important in regulating bone homeostasis and their manipulation may represent a new anabolic therapy for the treatment of disorders such as osteoporosis or non-union fractures.

Purpose: Since the neurotransmitter glutamate mediates nociceptive and pathological processes in arthritis, we have investigated how glutamate receptor and transporter expression varies with anatomical site or disease severity in subchondral bone of patients with osteoarthritis.

Methods and Results: Subchondral bone was sampled from tibial cuts derived from total knee arthroplasty (n=2, TKR, Kellgren Lawrence[KL] grade 3) and from tibial drill hole sites from high tibial osteotomy (n=2, HTO, KL grades 2 and 3) for osteoarthritis. RNA was extracted, reverse transcribed and RT-PCR performed for the housekeeping gene GAPDH, the glutamate transporter EAAT-1, and glutamate receptors (NR2A and KA1). Quantitative RT-PCR assessed differences in the expression of EAAT-1, a dominant negative splice variant called EAAT-1ex9skip and osteocalcin after normalisation to GAPDH.

Good quality RNA was obtained from bone cores removed from drill holes during HTO surgery, with GAPDH, EAAT-1, NR2A and KA1 expression detected. Osteocalcin expression was high indicating RNA was derived from osteoblasts and osteocytes, but did not vary with anatomical site or disease status. End-stage RT-PCR indicated differential expression of EAAT-1 between medial and lateral bone samples in total knee arthroplasty, however these differences were not significant by quantitative RT-PCR. In one patient, EAAT-1 expression was significantly reduced in the anterior zone versus the middle or posterior zones (ANOVA, p< 0.001). EAAT-1ex9skip represented a significant proportion of the total EAAT-1 mRNA expression in bone from TKR patients, but appeared less abundant in HTO samples.

Conclusion: We have shown for the first time that glutamate transporters and receptors are highly expressed in subchondral bone of patients with osteoarthritis and that EAAT-1 expression may vary with anatomical location and pathology. Activation of these receptors and transporters by the increased synovial fluid concentrations of glutamate that occur in arthritis may contribute to pathological changes and nociception.

Purpose of the study: Since the neurotransmitter glutamate mediates nociceptive and pathological processes in arthritis, we have investigated how glutamate receptor and transporter expression varies with anatomical site or disease severity in subchondral bone of patients with osteoarthritis.

Methods and results: Subchondral bone was sampled from tibial cuts derived from total knee arthroplasty (n=2, TKR, Kellgren Lawrence grade 3) and from tibial drill hole sites from high tibial osteotomy (n=2, HTO, KL grades 2 and 3) for osteoarthritis. RNA was extracted, reverse transcribed and RT-PCR performed for the housekeeping gene GAPDH, the glutamate transporter EAAT-1, and glutamate receptors (NR2A and KA1). Quantitative RT-PCR assessed differences in the expression of EAAT-1, a dominant negative splice variant called EAAT-1ex9skip and osteocalcin after nor-malisation to GAPDH.

Good quality RNA was obtained from bone cores removed from drill holes during HTO surgery, with GAPDH, EAAT-1, NR2A and KA1 expression detected. Osteocalcin expression was high indicating RNA was derived from osteoblasts and osteocytes, but did not vary with anatomical site or disease status. End-stage RT-PCR indicated differential expression of EAAT-1 between medial and lateral bone samples in total knee arthroplasty, however these differences were not significant by quantitative RT-PCR. In one patient, EAAT-1 expression was significantly reduced in the anterior zone versus the middle or posterior zones (ANOVA, p< 0.001). EAAT-1ex9skip represented a significant proportion of the total EAAT-1 mRNA expression in bone from TKR patients, but appeared less abundant in HTO samples.

Conclusion: We have shown for the first time that glutamate transporters and receptors are highly expressed in subchondral bone of patients with osteoarthritis and that EAAT-1 expression may vary with anatomical location and pathology. Activation of these receptors and transporters by the increased synovial fluid concentrations of glutamate that occur in arthritis may contribute to pathological changes and nociception.

Purpose of the study: Since glutamate can activate both nociceptive and pathological processes, we have investigated glutamate signalling in patients with painful and asymptomatic meniscal tears to determine which components are expressed, whether this varies in different anatomical regions of the meniscus and whether it is influenced by pain or degeneration.

Methods and results: Meniscus samples were obtained from two patients undergoing arthroscopic partial meniscal resection for chronic degenerate painful meniscal tears, from one patient with a torn painless meniscus and from the less affected compartment of the knee joint of three patients undergoing total knee arthroplasty. Menisci were dissected into anatomical regions (anterior horn, body, posterior horn, inner vascular, outer avascular), cryosectioned and RNA extracted. RNA was reverse transcribed and PCR performed for the housekeeping gene GAPDH and glutamate receptor subunits (NR2A, AMPA GluR3, KA1). Absolute quantitative RT-PCR assessed mRNA expression of glutamate transporters (EAAT-1, EAAT-1ex9skip) and type I collagen after normalisation to GAPDH or total RNA.

Human meniscus expressed GAPDH, type 1 collagen, EAAT-1, EAAT-1ex9skip, NR2A, AMPA GluR3 and KA1 mRNAs. Levels of EAAT-1 expression, normalised to GAPDH, did not differ between the inner and outer halves, or in the anterior, middle or posterior regions of menisci from the less affected compartments of arthritic knees. EAAT-1 expression appeared greater in the 2 painful, compared with the single non-painful meniscus. Interestingly, EAAT-1ex9skip was significantly more common within the outer zones (ANOVA, P=0.040) and in the posterior horns of the menisci (ANOVA, p=0.038).

Conclusion: We have shown for the first time that glutamate receptors and transporters are expressed in human meniscus providing a potential mechanism underlying the pathophysiology of pain associated with a torn meniscus. Our preliminary data indicate that EAAT-1 and EAAT-1ex9skip expression may vary with extent of damage and anatomical location in the human meniscus.