Effective NSAID treatment indicates that hyperprostaglandinism is affecting the clinical severity of childhood hypophosphatasia
© Girschick et al; licensee BioMed Central Ltd. 2006
Received: 24 February 2006
Accepted: 28 June 2006
Published: 28 June 2006
Hypophosphatasia (HP) is an inborn error of bone metabolism characterized by a genetic defect in the gene encoding the tissue-nonspecific alkaline phosphatase (TNSALP). There is a lack of knowledge as to how the variability and clinical severity of the HP phenotype (especially pain and walking impairment) are related to metabolic disturbances or impairments, subsequent to the molecular defect.
We analyzed the changes in clinical symptoms and the prostaglandin (PG) metabolism in response to treatment with non-steroidal anti-inflammatory drugs (NSAIDs) in six children affected by childhood HP. In addition, by exposing HP fibroblasts to pyridoxal phosphate and/or calcium pyrophosphate in vitro, we analyzed whether the alterations in PG levels are sequelae related to the metabolic defect.
Childhood HP patients, who often complain about pain in the lower limbs without evident fractures, have systemic hyperprostaglandinism. Symptomatic anti-inflammatory treatment with NSAIDs significantly improved pain-associated physical impairment. Calcium pyrophosphate, but not pyridoxal phosphate, induced cyclooxygenase-2 (COX-2) gene expression and PG production in HP and normal fibroblasts in vitro.
Clinical features of childhood HP related to pain in the lower legs may be, at least in part, sequelae related to elevated PG levels, secondary to the primary metabolic defect. Consequently, NSAID treatment does improve the clinical features of childhood HP.
Hypophosphatasia (HP) (MIM 241510) is an inborn error of bone metabolism, characterized by a genetic defect in the gene encoding the tissue-nonspecific alkaline phosphatase (TNSALP) [1–3]. The HP phenotype displays considerable clinical variability. Five major subtypes of HP (perinatal, infantile, childhood, adult and odontohypophosphatasia) have been described . There are biochemical  and molecular data [6–9] indicating that the severity of the molecular genetic alterations and subsequent changes in the levels of TNSALP are major determinants of the clinical phenotype. However, there is a lack of knowledge about how these features and the severity of the phenotype are influenced by the molecular defect and/or subsequent metabolic disturbances or sequelae. This seems of particular importance since metabolic products accumulating in HP patients, like pyrophosphate [10, 11], have the potential to provoke crystal-induced arthritis [12–16]. Since milder phenotypes, including the childhood form of HP, do not reduce the life span of the affected individuals significantly, there is considerably more time for such secondary metabolic phenomena to influence the HP phenotype. We have previously shown that childhood HP patients commonly complain about pain in the lower limbs, especially after physical exercise . In our previous study, symptomatic treatment of HP patients with one course of non-steroidal anti-inflammatory drugs (NSAID) resulted in a significant clinical improvement in physical activity and a reduction in pain-related complaints . So far, no long-term data on such a symptomatic treatment has been reported. As prostaglandins (PG) can induce appositional bone growth [18–20], we hypothesized that elevated PG values might be a reaction compensating for a mechanically incompetent bony structure, or it might be a sequel caused by the metabolic defect in HP . In order to avoid blocking PG synthesis for a long period of time, we now have investigated the effect of repeated NSAID treatments, each lasting 8 weeks, with an interval of four weeks in between.
We hypothesized that the systemic elevation of PG levels seen in our patients was due to pyrophosphate stimulation of connective tissue cells including fibroblasts; PG synthesis by skin and synovial non-hypophosphatasia fibroblasts has already been established . We, therefore, compared the prostaglandin synthesis occurring in normal as well as in HP fibroblasts exposed to pyridoxal phosphate and/or calcium pyrophosphate, two of the metabolic products that accumulate in hypophosphatasia. The goal of these analyses was to find out whether mechanisms secondary to TNSALP deficiency influence the clinical severity in patients diagnosed with childhood HP and whether pyrophosphate-stimulated HP fibroblasts are more prone to promote inflammation than healthy controls.
Patients and methods
TNSALP in leukocytes
2–18 nmol/min/mg protein
PEA in serum/plasma
PEA/Cr in urine
9–25 mmol/mol Cr
PLP in plasma
Bone metabolic profile
The following assays were performed: serum calcium and phosphate, intact parathyroid hormone (PTH) (1–84 amino acids), TNSALP and total serum alkaline phosphatase (AP). In urine, deoxypyridinoline/creatinine ratio, hydroxyproline/creatinine ratio, calcium/creatinine ratio and phosphate excretion were measured. Calcium concentrations were measured by flame atomic absorption, phosphate and creatinine concentrations were measured by standard colorimetric assays. Intact PTH was measured by a 2-site immunoradiometric assay (Nichols Institute, Bad Nauheim, Germany). Analysis of TNSALP and its substrates was performed by Y. S. Shin, MD (Munich, Germany) . Analysis of deoxypyridinoline/creatinine and hydroxyproline/creatinine levels in the urine was performed by K. Kruse, MD (†) and O. Hiort, MD (Lübeck, Germany) using enzyme immunoassays (Metra Biosystems Inc., Palo Alto, CA, USA) .
Prostaglandin profile in the urine during NSAID treatment
This study was funded by University funds only. The treatment was reviewed and approved by the ethics committee of the University of Würzburg, according to the principles of the Declaration of Helsinki.
Human fibroblasts were prepared from normal human skin (control), from skin of a HP patient and from dura mater of another HP patient (both patients underwent neurosurgery for release of increased intracranial pressure; these patients were not included in the biochemical analyses shown below). Tissue samples were prepared by trypsin enzymatic digestion, maintained in culture following a method described previously . The same amount of medium out of the same batch was used for the culture of the cells. Culture conditions between the samples were kept as comparable and constant as possible. The study was performed according to the principles of the Declaration of Helsinki. Isolation and culture of fibroblasts were approved by the ethics committee of the University of Würzburg, Germany. Fibroblasts were cultured in 10 cm2 culture plates (Nunc) using RPMI-1640 medium (Invitrogen), supplemented with 10% fetal calf serum (FCS) (Biochrom KG, Berlin, Germany) at 37°C in a 5% CO2 humidified incubator. They were used between passages 10 and 12. For all experiments, fibroblasts (1 × 105) were seeded in different culture plates and used at a confluent growth state. Normal and HP fibroblasts were stimulated with pyridoxal-5-phosphate (C8H10NO6P; Sigma Aldrich) at 0.5 micrograms/ml of culture medium for 48 hours in vitro. No pyridoxal-5-phosphate was present in the RPMI medium, however, pyridoxine hydrochloride was present at the same concentration both in the tested samples and controls. Stimulation with calcium pyrophosphate (Ca2P2O7; 200 microgram/ml culture medium, Aldrich) was performed for 48 hours in vitro. According the manufacturer's analysis, no calcium pyrophosphate was present in the RPMI culture medium. There was no available data on the content of both substances in the FCS supplement. Pyridoxal-5-phosphate (PLP) concentration was chosen according to the patients' blood levels. Calcium pyrophosphate (CPPD) concentration was chosen according to former publications [14, 16]. The experiments were done independently and repeated four times each. The concentrations of CPPD and PLP in the supplemented medium were the same in the different test samples as the same batch of supplemented medium was used.
Isolation of RNA and synthesis of first strand complementary DNA
Total cellular RNA was isolated from fibroblasts of each individual experimental set using the RNeasy mini kit (Qiagen) according to Qiagen's protocol. Approximately the same amount of cells were harvested from each sample. Contaminating genomic DNA was removed using RNase-free DNase (Qiagen). Aliquots of RNA from each experimental set were reversely transcribed into cDNA using Superscript 2 RNase H reverse transcriptase (RT) (Invitrogen). The cDNA content of the samples was adjusted by spectrophotometry. Two negative samples were prepared, one by adding only RT mix (RT+) without RNA and another lacking RT (RT-). Positive controls were run using total RNA isolated from tonsil tissue. The following reagents were added in each tube for the reverse transcription reaction mix: 8 μ l 5X buffer (Invitrogen), 4 μ l dTT (0.1 M) (Invitrogen), 2 μ l dNTP (Sigma), 2 μ l Superscript 2 RNase H-reverse transcriptase (Invitrogen) and 1 μ l Rnasin (Promega). The reaction mix along with RNA was initially incubated for 15 minutes at 70°C, followed by incubation at 42°C for 50 minutes and 70°C for 15 minutes, finally the tubes were kept on ice for 10 minutes. Tubes containing cDNA were stored at -70°C. Polymerase chain reaction (PCR) was performed to amplify cDNA of expressed genes of interest. The experiments were repeated four times each.
Quantification of cDNA in samples
For quantification of cDNA in the different samples, PCR amplification for the housekeeping gene β-actin was performed in triplicate at 25, 30 and 32 cycles each, after the initial adjustment of the cDNA content was done by spectrophotometry. RT-PCR products were analyzed on 1.8% agarose gels. The gel image was acquired by using a gel documentation unit (Bio-Rad). The intensities of the PCR product bands were analyzed by "quantity one" software (Bio-Rad). The mean counts of the intensities of these triplets in the linear range of PCR amplification were calculated and were used to calculate the amount of cDNA (in μ l) of each experimental sample that contained the same relative content of housekeeping gene cDNA, as previously described [21, 27, 28]. The calculated volume of cDNA was subjected to further PCR amplification of expressed genes of interest. The amplified PCR product was again analyzed by agarose gel electrophoresis.
Polymerase chain reaction (PCR)
The PCR was performed in a mastercycler (Eppendorf). The β-actin-adjusted amount of cDNA from the respective experimental sets was mixed with 20 μ l molecular biology grade water (Eppendorf), 2.5 μ l thermophilic DNA poly 10X buffer (Promega), 1.5μ l 25 mM MgCl2 (Promega), 0.5μ l 10 mM dNTP (Sigma) and 0.25 μ l of 50 pmol 3' and 5' primers (MWG, Ebersberg, Germany) in a total volume of 25 μ l. The primer pairs used in the experiment were as follows:
COX-1 sense 5' GGGCTGATGGTCTTAAATGC 3',
COX-1 antisense 5' GCCAGGAACACAACACTTTG 3',
COX-2 sense 5' CAGCACTTCACGCATCAGTT 3',
COX-2 antisense 5' CAGCAAACCGTAGATGCTCA3',
beta actin sense 5'GTCCTCTCCCAAGTCCACACA3',
beta actin antisense 5'CTGGTCTCAAGTCAGTGTACAGGTAA3'.
All cyclooxygenase (COXs) primer pairs were designed from the published human cDNA sequence data (NCBI Genbank) using primer design software (Primer 3', Whitehead Institute of Biomedical Research, USA). The PCR of the target genes was done at the following settings: 5 minutes initial denaturation at 95°C in the first cycle, followed by 1 minute denaturation at 94°C, 1 minute annealing at 60°C and 1 minute elongation at 72°C for all the primer sets. We used 38 cycles for each experiment to see the expression of our target sequences, as described previously [27, 28]. Experiments were repeated four times each. Intensities of the PCR product bands were calculated in a way that allowed comparison of different experiments. PCR band intensities of individual experiments (pyrophosphate and/or pyridoxal phosphate exposed and controls) were added and normalized to 100%. The percentages of the relative gene expression at a given time point in the four sets of experiments were added up, and the mean of relative cDNAexpression was calculated, as previously described [27, 28].
Enzyme linked immunosorbant assay (ELISA)
Mann Whitney's exact U-test was used for analyzing the changes in laboratory values (including PG data) and clinical status of patient #3 before and after treatment with meloxicam. Student's t-test was performed for comparison of the PG levels before and after treatment of all patients . Three-factorial analysis of variance (3-way ANOVA) was performed to describe differences in the COX-1, COX-2 gene expression, in addition to PG production of fibroblasts in vitro. P values smaller than 0.05 were considered significant, indicating a difference between the initial values and the follow-up data. Statistical analysis was performed by Dr. I. Haubitz.
Bone metabolism in patients
The diagnosis of childhood HP was based on clinical signs and confirmed in all patients by reduced serum alkaline phosphatase (AP) activity and elevated AP substrate, including phosphoethanolamine (PEA), levels in urine and elevated pyridoxal-5'-phosphate (PLP) in plasma . TNSALP, measured in leukocytes, was at the lowest detectable level in all patients (mean 0.15 nmol/min/mg protein, range 0–0.5 nmol/min/mg protein, normal range 2–18 nmol/min/mg protein) (Table 1). Genetic analysis of the TNSALP gene has been previously reported . Serum calcium concentrations were normal and phosphate levels were in the upper normal range. Interestingly, markers of bone turnover, including collagen crosslinks and hydroxyproline/creatinine levels in the urine, were unremarkable in all patients. The results obtained for patient #3 during the long-term prospective NSAID treatment with meloxicam are summarized in figures 2,3,4.
Evaluation of "on demand" NSAID treatment in all patients
Previously, we have shown that one single 5-week long NSAID treatment cycle reduced PG levels in the patients studied . During the follow-up, the anti-inflammatory treatment with naproxen was reinitiated in response to demand from the parents and patients because of clinical deterioration. Over periods of 4.25, 3.5, 6.5, 4, 4 and 3 years, 2, 2, 7, 3, 1 and 2 treatments, respectively, were demanded by the patients. Prior to NSAID treatment, the urinary PGE2 levels, reflecting renal PG production, were essentially within the normal range. No change was noted after 8 weeks of NSAID treatment (data not shown, paired 16 t-test, p = 0.1006). When systemic prostaglandin PGE-M production was evaluated (in 9 out of 15 treatment cycles) elevated levels (above the normal range) were detected before the therapy was started (figure 1). All PGE-M levels dropped into the normal range immediately after therapy (paired t-test, p = 0.0012) and did rise again after discontinuation of treatment (paired t-test, p = 0.0011) (Figure 1).
Long-term prospective evaluation of NSAID treatment
Based on our previous study of NSAID treatment , we wanted to prospectively evaluate NSAID treatment in order to analyze its effects on the clinical status, bone metabolism and bone mineral density. As indicated in material and methods, patient #3 was followed during four cycles of meloxicam treatment over one year. Meloxicam was selected after the patient developed pseudoporphyria after three naproxen treatments. Each treatment cycle lasted 8 weeks and was followed by a pause of 8 weeks.
Clinical evaluation of patient #3 using meloxicam as anti-inflammatory treatment
after 8 wks
overall well being, physician's global rating
overall well being, parents' global rating
fatigue after physical activity
patient wants to be carried
pain in rest
pain after physical activity
Laboratory analysis was done before and immediately after treatment, including a variety of parameters of bone metabolism (figure 3). There was no significant change in the concentration of calcium, phosphate, parathyroid hormone and creatinine in the serum, measured before and after treatment. Urinary excretion of hydroxyproline and deoxypyridinoline did not differ significantly before and after treatments. Functional analysis of the kidney including creatinine clearance, phosphate reabsorption (TRP) and phosphate transport maximum (TmPO4) were unchanged. All these values were within the normal range. However, serum alkaline phosphatase, measured before and after treatment, showed a significant increase from a mean of 39 U/l to 48.5 U/l (p = 0.032, Mann Whitney's exact U-Test) (figure 3). A selective analysis of TNSALP over time was not performed.
We also evaluated PG concentrations in the urine of patient #3 as indicated in figure 4. A significant reduction of systemic PGE-M levels, which were elevated in four out of five pre-treatment periods, was noted (figure 4) (Mann Whitney's exact U-Test: p = 0.016). After treatment, the PGE-M concentration was in the normal range. Renal PGE2production, which was not elevated before treatment, did not change significantly during the treatment cycles (Mann Whitney's exact U-Test: p = 0.3).
In vitro cyclooxygenase gene expression in normal and HP fibroblasts exposed to pyridoxal phosphate (PLP) and calcium pyrophosphate (CPPD)
In order to evaluate a potential stimulation of normal and HP fibroblasts with PLP and/or CPPD in vitro, COX-1 and COX-2 gene expression was analyzed in these cells. PGE2 production was assessed by measuring PGE2 concentration in the supernatant, using the ELISA technique. COX-1 gene expression did not show a consistent change when CPPD was added either to normal or to HP skin fibroblasts, or HP fibroblasts isolated from the dura mater (3-way ANOVA, p = 0.16). In addition, no significant change was noted when PLP was added to the cultures, compared to the non-treated cells in each group (3-way ANOVA, p = 0.93). When PLP together with CPPD were added, a slight statistically insignificant increase in gene expression was noted in normal and HP skin fibroblasts. However, a reduction in the gene expression was found in HP dura mater fibroblasts. None of these changes in COX-1 expression, however, were statistically significant.
In normal fibroblasts, the addition of PLP did not alter COX-2 gene expression, whereas in the two HP fibroblast types, a slight increase of COX-2 production was noted (3-way ANOVA, p = 0.71) (figure 5). When CPPD was added to previously non-treated normal or HP fibroblasts, a significant increase of COX-2 gene expression was noted in each fibroblast type (3-way ANOVA, p = 0.0067). When PLP together with CPPD was added, an increase in COX-2 expression was noted in normal and HP skin fibroblasts, which was comparable to the stimulation with CPPD alone (figure 5). When these latter two sets were compared, no significant difference was noted (3-way ANOVA, p = 0.62), indicating that PLP did not contribute significantly to the COX-2 gene induction and PG production.
In vitro prostaglandin production by normal and HP fibroblasts exposed to pyridoxal phosphate (PLP) and calcium pyrophosphate (CPPD) crystals
When PG concentrations were measured in the culture supernatant, a significant increase of PGE2 production was noted after calcium pyrophosphate was added to the cultures of either normal fibroblasts, HP skin or HP dura mater fibroblasts (3-way ANOVA, p = 0.011). No effect was observed when PLP was added to the cultures alone (3-way ANOVA, p = 0.88). No synergistic effect was noted when CPPD was added together with PLP (3-way ANOVA, p = 0.9) (figure 6).
Each of the childhood HP patients experienced a significant improvement of their physical status when treated with NSAIDs. This clinical effect was striking and could be noted in a "treatment on demand" setting involving all patients, as well as in repetitive treatments involving one patient. The latter was treated prospectively with the anti-inflammatory medication meloxicam for one year, including treatment pauses. Treated patients reported that the positive effect of the therapy was already noticeable within a few days after starting the NSAID treatment. After drug withdrawal, the positive effect on physical activity persisted for 3–6 weeks. Patient #3 did not experience an effect lasting longer than 3–4 weeks after NSAID therapy. This was consistent with a rise in PG levels before each treatment cycle. An excessive rebound of PG synthesis after cessation of COX blockade was not noted. An extensive concomitant laboratory analysis did not reveal changes in bone metabolism, kidney function and parathyroid function related to NSAID treatment. However, a significant increase of alkaline phosphatase activity in the serum was noted. So far, it is not clear whether this means an actual increase in TNSALP function or concentration, or whether it is caused by a variation in sample results measured at different time points. Nevertheless, it is important to analyze this particular effect in more detail in the future, because induction of TNSALP gene expression might be a therapeutic target in HP.
The pattern in mineralization obtained over the whole period of four years through bone mineral density measurements in the patient 3 undergoing repetitive treatments did not differ from that obtained from patients treated on an "on demand" basis for a limited time only. The improvements in the physical activity noted by the parents of this child during the treatment cycles, however, did not seem to influence the dynamics of bone mineral density to a noticeable extent during the follow-up of 4 years. Initially, we had feared that a PG synthesis blockade could even worsen bone mineralization, since PGs may have an inhibitory effect on bone formation rather than a positive effect. One can only speculate whether a long-term continuous anti-inflammatory and analgesic therapy can lead to an improvement or, at least, to a stabilization of the bony structure, secondary to an improvement of the physical activity.
In vitro analysis showed that normal and HP fibroblasts are equally sensitive to calcium pyrophosphate exposure. A significant increase of COX-2 gene expression and subsequent increase of PGE2 concentrations in the supernatant were noted in fibroblasts isolated from normal skin, HP skin and HP dura mater, after addition of calcium pyrophosphate but not pyridoxal phosphate. There was no HP-specific effect, showing that HP fibroblasts are particularly oversensitive to the exposure of calcium pyrophosphate. Thus, this study provides evidence that the elevation of PGs in HP might be based on the impaired clearance of calcium pyrophosphate resulting from alkaline phosphatase deficiency. Our data suggest that PGs can reduce the physical activity of childhood HP patients . Physical inactivity could lead to a further impairment of bone mineralization, especially in the background of this genetic defect. Strenuous physical work carried out by patient #6 might have contributed to an increase of bone formation, which reached physiological levels. In contrast, patient #3, who was treated with NSAID for a total duration of 32 weeks, showed no improvement in bone mineral density. Thus, physical activity seems to have a more positive impact on mineralization in HP than normalization of potentially elevated PGs alone. Further studies are needed to understand the secondary metabolic phenomena associated with this inborn defect of bone metabolism and affecting the severity of the clinical phenotype.
non-steroidal anti-inflammatory drug
polymerase chain reaction
systemic index metabolite of prostaglandin E2
tubular transport maximum of phosphate
tissue non-specific alkaline phosphatase
tubular absorption of phosphate
We are indebted to Martina Rübig for her excellent technical assistance.
- Greenberg CR, Evans JA, McKendry-Smith S, Redekopp S, Haworth JC, Mulivor R, Chodirker BN: Infantile hypophosphatasia: localization within chromosome region 1p36.1-34 and prenatal diagnosis using linked DNA markers. Am J Hum Genet. 1990, 46 (2): 286-292.PubMed CentralPubMed
- Henthorn PS, Whyte MP: Missense mutations of the tissue-nonspecific alkaline phosphatase gene in hypophosphatasia. Clin Chem. 1992, 38 (12): 2501-2505.PubMed
- Henthorn PS, Raducha M, Fedde KN, Lafferty MA, Whyte MP: Different missense mutations at the tissue-nonspecific alkaline phosphatase gene locus in autosomal recessively inherited forms of mild and severe hypophosphatasia. Proc Natl Acad Sci U S A. 1992, 89 (20): 9924-9928. 10.1073/pnas.89.20.9924.PubMed CentralView ArticlePubMed
- Whyte MP: Heritable metabolic and dysplastic bone diseases. Endocrinol Metab Clin North Am. 1990, 19 (1): 133-173.PubMed
- Whyte MP, Walkenhorst DA, Fedde KN, Henthorn PS, Hill CS: Hypophosphatasia: levels of bone alkaline phosphatase immunoreactivity in serum reflect disease severity. J Clin Endocrinol Metab. 1996, 81 (6): 2142-2148. 10.1210/jc.81.6.2142.PubMed
- Weiss MJ, Cole DE, Ray K, Whyte MP, Lafferty MA, Mulivor R, Harris H: First identification of a gene defect for hypophosphatasia: evidence that alkaline phosphatase acts in skeletal mineralization. Connect Tissue Res. 1989, 21 (1–4): 99-104. discussion 104–106.View ArticlePubMed
- Hu CC, King DL, Thomas HF, Simmer JP: A clinical and research protocol for characterizing patients with hypophosphatasia. Pediatr Dent. 1996, 18 (1): 17-23.PubMed
- Zurutuza L, Muller F, Gibrat JF, Taillandier A, Simon-Bouy B, Serre JL, Mornet E: Correlations of genotype and phenotype in hypophosphatasia. Hum Mol Genet. 1999, 8 (6): 1039-1046. 10.1093/hmg/8.6.1039.View ArticlePubMed
- Orimo H, Girschick HJ, Goseki-Sone M, Ito M, Oda K, Shimada T: Mutational analysis and functional correlation with phenotype in German patients with childhood-type hypophosphatasia. J Bone Miner Res. 2001, 16 (12): 2313-2319. 10.1359/jbmr.2001.16.12.2313.View ArticlePubMed
- Sorensen SA, Flodgaard H, Sorensen E: Serum alkaline phosphatase, serum pyrophosphatase, phosphorylethanolamine and inorganic pyrophosphate in plasma and urine. A genetic and clinical study of hypophosphatasia. Monogr Hum Genet. 1978, 10: 66-69.View ArticlePubMed
- Whyte MP, Mahuren JD, Vrabel LA, Coburn SP: Markedly increased circulating pyridoxal-5'-phosphate levels in hypophosphatasia. Alkaline phosphatase acts in vitamin B6 metabolism. J Clin Invest. 1985, 76 (2): 752-756.PubMed CentralView ArticlePubMed
- Brandt KD, Schumacher HR Jr: Osteoarthritis and crystal deposition diseases. Curr Opin Rheumatol. 1993, 5 (4): 483-486.View ArticlePubMed
- Doherty M, Chuck A, Hosking D, Hamilton E: Inorganic pyrophosphate in metabolic diseases predisposing to calcium pyrophosphate dihydrate crystal deposition. Arthritis Rheum. 1991, 34 (10): 1297-1303.View ArticlePubMed
- Dayer JM, Evequoz V, Zavadil-Grob C, Grynpas MD, Cheng PT, Schnyder J, Trechsel U, Fleisch H: Effect of synthetic calcium pyrophosphate and hydroxyapatite crystals on the interaction of human blood mononuclear cells with chondrocytes, synovial cells, and fibroblasts. Arthritis Rheum. 1987, 30 (12): 1372-1381.View ArticlePubMed
- Cheung HS, Halverson PB, McCarty DJ: Release of collagenase, neutral protease, and prostaglandins from cultured mammalian synovial cells by hydroxyapatite and calcium pyrophosphate dihydrate crystals. Arthritis Rheum. 1981, 24 (11): 1338-1344.View ArticlePubMed
- McCarty DJ, Cheung HS: Prostaglandin (PG) E2 generation by cultured canine synovial fibroblasts exposed to microcrystals containing calcium. Ann Rheum Dis. 1985, 44 (5): 316-320.PubMed CentralView ArticlePubMed
- Girschick HJ, Seyberth HW, Huppertz HI: Treatment of childhood hypophosphatasia with nonsteroidal antiinflammatory drugs. Bone. 1999, 25 (5): 603-607. 10.1016/S8756-3282(99)00203-3.View ArticlePubMed
- Raisz LG: Potential impact of selective cyclooxygenase-2 inhibitors on bone metabolism in health and disease. Am J Med. 2001, 110 (Suppl 3A): 43S-45S. 10.1016/S0002-9343(00)00684-7.View ArticlePubMed
- Raisz LG: Prostaglandins and bone: physiology and pathophysiology. Osteoarthritis Cartilage. 1999, 7 (4): 419-421. 10.1053/joca.1998.0230.View ArticlePubMed
- Williams JL: Periosteal hyperostosis resulting from prostaglandin therapy. Eur J Radiol. 1986, 6 (3): 231-232.PubMed
- Girschick HJ, Schneider P, Kruse K, Huppertz HI: Bone metabolism and bone mineral density in childhood hypophosphatasia. Bone. 1999, 25 (3): 361-367. 10.1016/S8756-3282(99)00164-7.View ArticlePubMed
- Shin YS, Rasshofer R, Friedrich B, Endres W: Pyridoxal-5'-phosphate determination by a sensitive micromethod in human blood, urine and tissues; its relation to cystathioninuria in neuroblastoma and biliary atresia. Clin Chim Acta. 1983, 127 (1): 77-85. 10.1016/0009-8981(83)90077-3.View ArticlePubMed
- Rauch F, Seibel M, Woitge H, Kruse K, Schonau E: Increased urinary excretion of collagen crosslinks in girls with Ullrich-Turner syndrome. Acta Paediatr. 1995, 84 (1): 66-69.View ArticlePubMed
- Schad SG, Hamm H, Girschick HJ: [Bullae, erosions and scars in the face of 2 girls with idiopathic arthritis. Naproxen-induced pseudoporphyria]. Hautarzt. 2002, 53 (1): 51-53. 10.1007/s105-002-8049-2.View ArticlePubMed
- Seyberth HW, Koniger SJ, Rascher W, Kuhl PG, Schweer H: Role of prostaglandins in hyperprostaglandin E syndrome and in selected renal tubular disorders. Pediatr Nephrol. 1987, 1 (3): 491-497. 10.1007/BF00849259.View ArticlePubMed
- Huppertz HI: How could infectious agents hide in synovial cells? Possible mechanisms of persistent viral infection in a model for the etiopathogenesis of chronic arthritis. Rheumatol Int. 1994, 14 (2): 71-75. 10.1007/BF00300250.View ArticlePubMed
- Girschick HJ, Grammer AC, Nanki T, Mayo M, Lipsky PE: RAG1 and RAG2 expression by B cell subsets from human tonsil and peripheral blood. J Immunol. 2001, 166 (1): 377-386.View ArticlePubMed
- Singh SK, Morbach H, Nanki T, Faber C, Baar V, Wirsing A, Girschick HJ: Differential expression of matrix metalloproteinases and cyclooxygenases in synovial cells infected with different Borrelia burgdorferi isolates. Inflamm Res. 2004, 53: 689-696. 10.1007/s00011-004-1313-x.View ArticlePubMed
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.