Response to Inflammatory Stimuli Trafficking and MMP-9 Secretion in Arthritis Severity, Modulates Lysosomal -Glucuronidase, a Re

Home , Enzyme assay

β-Glucuronidase, a Regulator of Lyme Arthritis Severity, Modulates Lysosomal Trafficking and MMP-9 Secretion in Response to Inflammatory Stimuli This information is current as of October 5, 2021. Kenneth K. C. Bramwell, Kelton Mock, Ying Ma, John H. Weis, Cory Teuscher and Janis J. Weis J Immunol 2015; 195:1647-1656; Prepublished online 13 July 2015;

doi: 10.4049/jimmunol.1500212 Downloaded from http://www.jimmunol.org/content/195/4/1647

References This article cites 40 articles, 9 of which you can access for free at: http://www.jimmunol.org/content/195/4/1647.full#ref-list-1 http://www.jimmunol.org/

Why The JI? Submit online.

• Rapid Reviews! 30 days* from submission to initial decision

• No Triage! Every submission reviewed by practicing scientists

by guest on October 5, 2021 • Fast Publication! 4 weeks from acceptance to publication

*average

Subscription Information about subscribing to The Journal of Immunology is online at: http://jimmunol.org/subscription Permissions Submit copyright permission requests at: http://www.aai.org/About/Publications/JI/copyright.html Email Alerts Receive free email-alerts when new articles cite this article. Sign up at: http://jimmunol.org/alerts

The Journal of Immunology is published twice each month by The American Association of Immunologists, Inc., 1451 Rockville Pike, Suite 650, Rockville, MD 20852 Copyright © 2015 by The American Association of Immunologists, Inc. All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. The Journal of Immunology b-Glucuronidase, a Regulator of Lyme Arthritis Severity, Modulates Lysosomal Trafﬁcking and MMP-9 Secretion in Response to Inﬂammatory Stimuli

Kenneth K. C. Bramwell,* Kelton Mock,† Ying Ma,* John H. Weis,* Cory Teuscher,‡ and Janis J. Weis*

The lysosomal enzyme b-glucuronidase (Gusb) is a key regulator of Lyme-associated and K/B3N-induced arthritis severity. The luminal enzymes present in lysosomes provide essential catabolic functions for the homeostatic degradation of a variety of macromolecules. In addition to this essential catabolic function, lysosomes play important roles in the inflammatory response following infection. Secretory lysosomes and related vesicles can participate in the inflammatory response through fusion with the plasma membrane and release of bioactive contents into the extracellular milieu. In this study, we show that GUSB hypomorphism potentiates lysosomal exocytosis following inflammatory stimulation. This leads to elevated secretion of lysosomal contents, Downloaded from including glycosaminoglycans, lysosomal hydrolases, and matrix metalloproteinase 9, a known modulator of Lyme arthritis severity. This mechanistic insight led us to test the efficacy of rapamycin, a drug known to suppress lysosomal exocytosis. Both Lyme and K/B3N-associated arthritis were suppressed by this treatment concurrent with reduced lysosomal release. The Journal of Immunology, 2015, 195: 1647–1656.

yme disease is the most prevalent arthropod-borne illness consistent genetic differences in Lyme arthritis severity. They http://www.jimmunol.org/ in the United States with .300,000 estimated cases per observed that C3H/HeN (C3H) mice develop severe Lyme ar- L year (1). Lyme disease infection occurs following the bite thritis, whereas C57BL/6N (B6) mice experience milder inflam- of a tick infected with the spirochete Borrelia burgdorferi (2). In mation and pathology, representing opposite ends of the disease a majority of cases, infection leads to the development of an spectrum. This observation allowed the influence of underlying erythema migrans bull’s-eye rash at the site of the tick bite, which host genetic differences to be investigated in a systematic way. radiates outward along with dissemination of bacteria. Patients Through a forward genetic approach, we recently demonstrated experience a wide spectrum of disease symptoms following infec- that the lysosomal enzyme b-glucuronidase (Gusb) is a key reg- tion, which are thought to reflect differences in the infectious bac- ulator of Lyme arthritis severity (6). This result was extended to by guest on October 5, 2021 teria as well as in the patients’ own unique genetic risk factors. the K/B3N serum transfer model of rheumatoid arthritis, indi- Lyme arthritis is the most prevalent symptom of disseminated in- cating a common role for Gusb in these two models of arthritis. fection, observed in 30–60% of untreated patients (3). Although The Gusb gene in C3H mice contains a single nucleotide poly- early treatment with appropriate antibiotic therapy often effectively morphism that encodes a T87I amino acid substitution and leads to prevents the manifestation of later symptoms associated with dis- a partial reduction in GUSB enzymatic activity. GUSB is a critical seminated infection, up to 10% of treated patients experience re- enzyme in the lysosomal pathway used to degrade and recycle gly- current arthritis symptoms that may persist for months or years (4). cosaminoglycans (GAGs), and individuals with very severe GUSB Wild mice and other rodents serve a key role in the enzootic deficiencies exhibit spontaneous lysosomal storage of partially de- cycle of B. burgdorferi in nature (2). Barthold et al. (5) made the graded GAGs and severe joint/skeletal abnormalities (7). C3H mice pivotal observation that inbred strains of laboratory mice exhibit do not develop spontaneous lysosome storage symptoms until ad- vanced age, but GUSB hypomorphism does drive substantial arthritogenesis in young mice following challenge with an inflammatory *Division of Microbiology and Immunology, Department of Pathology, University of trigger (6). The cellular and molecular mechanisms underlying the Utah, Salt Lake City, UT 84112; †University of Puget Sound, Tacoma, WA 98416; and ‡Department of Medicine, University of Vermont, Burlington, VT 05405 observed joint pathogenesis are not fully understood, but the sub- Received for publication January 28, 2015. Accepted for publication June 17, 2015. cellular localization of GUSB to the lysosome and its indispensable This work was supported by National Institutes of Health/National Institute of Allergy enzymatic role implicate alterations in lysosomal function or traf- and Infectious Disease Grants T32AI055434 (to K.K.C.B.) and AI32223 (to J.J.W.), ficking as strong candidates for additional investigation. National Institutes of Health/National Institute of Arthritis and Musculoskeletal and Skin Diseases Grant AR43521 (to C.T. and J.J.W.), and Arthritis Foundation Grant 6133 (to K.K.C.B.). The DNA/Peptide Core is supported by Cancer Center Support Materials and Methods Grant P30 CA04201. Mice Address correspondence and reprint requests to Dr. Janis J. Weis, Division of Mi- crobiology and Immunology, Department of Pathology, University of Utah, 15 North All mice used in this study were maintained in a pathogen-free facility and Medical Drive, Salt Lake City, UT 84112. E-mail address: [email protected] cared for in accordance with protocols approved by the University of Utah b Institutional Animal Care and Use Committee. B6 and C3H mice were Abbreviations used in this article: B6, C57BL/6N; BGAL, -galactosidase; BMDM, obtained from the National Cancer Institute and Charles River Laboratories, bone marrow–derived macrophage; C3H, C3H/HeN; GAG, glycosaminoglycan; Null Het Gusb, b-glucuronidase; MEF, mouse embryonic fibroblast; MMP-9, matrix metal- respectively. Gusb homozygous and heterozygous (Gusb ) mice on loproteinase 9; mTOR, mammalian target of rapamycin; qRT-PCR, quantitative RT- a B6 genetic background were originally obtained from The Jackson h h PCR; Thio, Brewer modified thioglycollate medium. Laboratory, as described (6). B6.C3H-Gusb (B6-Gusb ) congenic mice were developed as previously described (6) and isolate a C3H-derived Copyright Ó 2015 by The American Association of Immunologists, Inc. 0022-1767/15/$25.00 locus on Chromosome 5 from 129.0 to 130.5 Mb, containing only one www.jimmunol.org/cgi/doi/10.4049/jimmunol.1500212 1648 Gusb MODULATES LYSOSOMAL TRAFFICKING AND MMP-9 SECRETION coding polymorphism that is located in the Gusb gene, on an otherwise length of 460 nm. Units were calculated by comparison against a standard uniform B6 genetic background. curve prepared using free 4-methylumbelliferone (Sigma-Aldrich). Results are expressed as arbitrary units, with the average value of B6 unstimulated Microscopy control samples within each assay run set at a baseline of 100. Alcian blue–stained sections were visualized on an Olympus BX41 clinical Gelatin zymography microscope (Olympus America) using 34, 310, or 340 total magnification. Images were recorded with an Olympus DP72 camera and prepared A total of 20 ml cell-free BMDM supernatant was added to 20 ml Zy- using Olympus cellSens digital imaging software (Olympus America). mogram Sample Buffer (Bio-Rad) and loaded onto 10% polyacrylamide Electron microscopy images were obtained using a JEOL JEM-1400 Plus gels embedded with gelatin (Bio-Rad). Electrophoresis buffer was pre- Transmission Electron Microscope (JEOL). pared from a 103 stock solution (15 g Tris base, 72 g glycine, and 5 g NaDodSO4 dissolved in 500 ml Milli-Q water). Zymography gels were LAMP-1 flow cytometry assays electrophoresed at 100 V for 1 h. After electrophoresis, gels were removed 3 Unelicited resident peritoneal cells were harvested from B6, C3H, and from the cassette and incubated in 1 Zymogram Renaturation Buffer GusbNull littermates as described (8). Cells contained in this naive peritoneal (Bio-Rad) at room temperature on a rocker plate for 30 min, followed by exudate were resuspended at a concentration of 5 3 105/ml in serum- overnight incubation at 37˚C in Zymography Development Buffer (Bio- replacement medium (RPMI 1640 medium [Invitrogen] supplemented with Rad). Gels were then stained with 2.5% Coomassie Blue R-250 (Bio-Rad) dissolved in a solution of 50% methanol, 10% glacial acetic acid, and 40% 1% L-glutamine and 1% Nutridoma SP [Roche]) (Control) or serum- Milli-Q water for 30 min. Gels were destained using 50% methanol, 10% replacement medium supplemented with 3% Brewer modified thioglycollate medium (3% Thio; Difco). A total of 2 ml cell suspension was glacial acetic acid, and 40% Milli-Q water and then imaged. Density aliquoted into 12 3 75 mm Falcon tubes (BD Biosciences), capped, and analysis of the bands was performed using Image J (National Institutes of incubated in a 37˚C water bath for 2 h. Samples were then washed twice Health). The appropriate position of the MMP-9 zymography band was with FACS buffer and stained with 7-aminoactinomycin D for viability dis- verified using recombinant mouse MMP-9 (AnaSpec) (data not shown). crimination and LAMP-1 PE-conjugated Ab (clone 1D4B). For lineage dis- ELISA Downloaded from crimination, peritoneal exudate was stained with DAPI for viability discrimination and a mixture of mAbs against LAMP-1 Alexa Fluor 488 MMP-9 ELISA was performed according to the manufacturer’s instructions (clone 1D4B), F4/80 APC (clone BM8), and CD19 BV605 (clone 6D5). All using a Quantikine ELISA Mouse Total MMP-9 kit (R&D Systems). Briefly, Abs were sourced from BioLegend and used at a 1:200 dilution. Sample data microplate strips were placed into a 96-well microplate and filled with 50 ml were collected on an FACSCanto II (BD Biosciences) and analyzed using assay diluent. Then, 50 ml cell-free BMDM supernatants, standards, or FlowJo v. 9.4.11 software (Tree Star). Live single cells were selected by control solutions were added, mixed, and incubated for 2 h at room tem- gating all samples on forward scatter height versus forward scatter width to perature. Each well was then aspirated and washed four times with wash http://www.jimmunol.org/ exclude doublets and then by subsequently gating out any cells staining buffer. A total of 100 ml Mouse Total MMP-9 Conjugate was added to each positively for 7-aminoactinomycin D or DAPI, respectively (data not shown). well, covered, and incubated for 2 h at room temperature. The plate was again aspirated and washed four times. A total of 100 ml substrate solution Culture of bone marrow–derived macrophages was added, the plate was covered and incubated at room temperature for m Rear ankle joints were measured at the time of infection and at 4 wk after 30 min in the dark. A total of 100 l stop solution was added to each well, infection by using a metric caliper, as described (9). BMDM growth medium and the plate was read using a BioTek Synergy HT microplate reader (BioTek) consisted of RPMI 1640 medium (Invitrogen) supplemented with 30% L929 at an absorbance of 450 nm and a wavelength correction set to 540 nm. conditioned medium and 20% horse serum (HyClone). Strain-specific Rapamycin treatment of mice BMDM were harvested, resuspended in serum-replacement medium at a density of 6 3 105/ml, and 500 ml aliquots were replated into 24-well Rapamycin(LCLaboratories)wasdissolvedinEtOHtopreparea100mg/ml plates and cultured overnight. stock solution. This was diluted to 1.2 mg/ml into a vehicle solution of 13 PBS, by guest on October 5, 2021 5% PEG-400, and 5% Tween-20. B6-Gusbh mice were given daily i.p. Culture of B. burgdorferi and BMDM coincubation injections of either 8 mg/kg rapamycin or an equivalent volume of vehicle, N40 isolate B. burgdorferi cells (provided by Stephen Barthold, University of beginning 2 d prior to the initiation of the experimental arthritis protocols and California Davis, Davis, CA) were cultured in Barbour-Stoenner-Kelly II continuing throughout the course of the experiment with the final injection medium containing 6% rabbit serum (Sigma-Aldrich). Live B. burgdorferi given the day prior to sacrifice. were visualized and counted using a Petroff-Houser counting chamber and K/B3N serum transfer experiments a dark-field condenser. An aliquot of B. burgdorferi was centrifuged at 6000 3 g on a tabletop centrifuge for 6 min, resuspended in room temperature 13 Experiments were carried out as described (6). Briefly, rear ankle joints of mice PBS, centrifuged again, and resuspended in serum-replacement medium at between 6 and 7 wk of age were measured with a metric caliper prior to a high concentration of 7.4 3 106/ml or a low concentration of 7.4 3 105/ml. treatment. A total of 100 mlK/B3N serum was administered in two i.p. The 24-well plates of strain-specific BMDM were removed from the incu- injections, with one on day 0 and one on day 2. Ankle swelling was determined bator, and the serum-replacement culture medium was aspirated and replaced by caliper measurements on days 5 and 7. After the final day 7 measurements, with 500 ml freshly prepared serum-replacement medium, low-concentration serum was collected for enzyme assays, the most swollen rear ankle joint was B. burgdorferi medium, or high-concentration B. burgdorferi medium and harvested for histological analysis, and the less swollen joint and the spleen returned to the incubator. The high and low concentrations used produce were excised, immersed in RNA Later (Qiagen), and stored at 280˚C. a multiplicity of infection of ∼25 and 2.5, respectively. Twenty-four hours later, supernatants were harvested into individual 1.7-ml microcentrifuge B. burgdorferi infection of mice tubes and centrifuged at 6000 3 g for 6 min. A total of 400 ml resulting cell- Mice between 6 and 7 wk of age were infected by intradermal injection with free supernatants was then transferred into a 96-well 500 ml round-bottom 2 3 104 bacteria of the B. burgdorferi N40 isolate. Sonicated B. burgdorferi assay plate (Axygen). To obtain cell extracts, cells were washed once with used for in vitro stimulation was prepared as described (9). 13 PBS and then incubated in extraction buffer (50 mmol NaPO4 [pH 7], 10 mmol BME, 10 mmol EDTA, 0.1% sarcosyl, and 0.1% Triton X-100) Lyme arthritis analysis on a rocker plate at 4˚C for 30 min. Extracts were then homogenized by pipetting up and down, and 400 ml was transferred into a 96-well assay plate. Rear ankle joints were measured at the time of infection and at 4 wk postin- fection by using a metric caliper, as described (9). Measurements of the thickest b-Galactosidase enzyme assay anteroposterior portion of the ankle with the joint extended were taken and are reported as the change in ankle swelling over time. For histological evaluation, 4-Methylumbelliferyl b-D-galactopyranoside (Marker Gene) was used as b tissues were fixed in 10% neutral buffered formalin, decalcified, embedded in a fluorogenic substrate to measure -galactosidase (BGAL) enzymatic m activity. A total of 10 ml sample (serum, cell extract, or supernatant) was paraffin, and cut into 3- m–thick sections; sections were mounted onto glass slides and stained with H&E or Alcian blue as described (6). incubated with 1 mmol 4-methylumbelliferyl b-D-galactopyranoside in a total volume of 50 ml assay buffer (100 mmol sodium citrate [pH 4.4]) Isolation of RNA and quantitative RT-PCR for 1 h at 37˚C in a 96-well plate (Costar). A total of 150 ml stop buffer (200 mmol sodium carbonate) was then added, and samples were analyzed Total RNA was recovered from homogenized tissue using TRIzol reagent on a BioTek Synergy HT microplate reader (BioTek). Fluorescence was (Invitrogen). 5 mg total RNA was reverse transcribed using random primers measured with an excitation wavelength of 360 nm and emission wave- and Moloney murine leukemia virus RT (Invitrogen). Transcripts were The Journal of Immunology 1649 quantified using LightCycler Sybr Plus master mix and a LC-480 PCR mate control mice. Following B. burgdorferi infection, GusbHet, system (Roche). The primers used were as follows: b-actin forward 59- B6-Gusbh, and GusbNull mice develop mild, moderate, and max- GTAACAATGCCATGTTCAAT-39 and reverse 59-CTCCATCGTGGGC- 9 9 9 imal arthritis, respectively (6). The Alcian blue positivity in joints CGCTCTAG-3 ; and Glb1 forward 5 -GTGGATAAATGGCTGGCAGT-3 Het and reverse 59-TACCCAGATGGTAGCGGAAG-39. from both sham and infected Gusb mice is primarily restricted to the articular surfaces of the cartilage and within the bones of the Results leg, ankle, and foot (Fig. 1). Ankle joints from infected B6-Gusbh We previously identified an increased extracellular GAG deposi- mice exhibit extracellular GAG deposition in areas associated tion, visualized histochemically with Alcian blue staining, and with histopathology, and this localized deposition is exaggerated localized inflammation following B. burgdorferi infection in se- further in GusbNull mice. Because the cellularity of the cranial– verely Gusb-deficient mice on a B6 genetic background (GusbNull) tibial tendon sheath is so widely variable between sham and and several Gusb hypomorphic strains (6, 10). The previously infected mice, we attempted to focus high-magnification (340) described runted growth and spontaneous rheumatologic abnor- images on an adjacent area that is densely cellular in all samples. malities of GusbNull mice suggest that the presence of excess The extracellular deposition of these GAGs is most apparent at partially degraded GAGs may be responsible for a shared disease this high magnification. Although GusbNull mice are expected to process underlying both pathologies (11). Additionally, the results spontaneously accumulate excess GAGs even in the absence of of radiation chimera studies indicated that the Gusb genotype of infection (11), we found that joints from sham-injected GusbNull joint resident cells determines the severity of arthritis, suggesting mice exhibit minimal extracellular GAG staining. To quantify that GAG released by these cells could contribute to localized intracellular GAG storage, BMDM from GusbHet and GusbNull inflammation. To investigate this, we compared the Alcian blue– mice were prepared and visualized by transmission electron mi- positive GAG staining patterns of joints from sham-injected and croscopy (Fig. 2). As expected, GusbNull cells exhibited many Downloaded from B. burgdorferi–infected B6-Gusbh, GusbNull, and GusbHet litter- electron sparse storage vacuoles of irregular size and shape, http://www.jimmunol.org/ by guest on October 5, 2021

FIGURE 1. GUSB insufficiency is associated with localized extracellular accumulation of GAGs following infection. Representative light microscopy images of Alcian blue–stained rear ankle joint sections from sham injected (Sham) or B. burgdorferi–infected (+ B.b.) mice. In each 34 total magnification image, the cranial–tibial tendon is identified with a black arrowhead. Black rectangles indicate the area selected for higher magnification in adjacent images, with total magnifications 310 and 340, respectively. Notably, GusbNull joints do not exhibit ex- tensive extracellular GAG deposition in the absence of infection. 1650 Gusb MODULATES LYSOSOMAL TRAFFICKING AND MMP-9 SECRETION Downloaded from http://www.jimmunol.org/

FIGURE 2. Severe GUSB deﬁciency is associated with intracellular storage vacuolation under basal conditions. Representative transmission electron micrographs of BMDM produced from GusbHet or GusbNull littermates.

GusbNull cells exhibit many electron sparse storage vacuoles of irregular size by guest on October 5, 2021 and shape, consistent with a lysosome storage disease phenotype. consistent with a lysosome storage disease phenotype. Together, these data suggest that although Gusb deficiency leads to intracellular accumulation of excess GAG, which may have an independent role in arthritogenesis that deserves further study, Gusb appears to play a distinct additional role in intracellular vesicular trafficking in response to inflammatory stimuli. The role of Gusb deficiency in lysosomal trafficking was investigated by measuring the amount of the lysosomal integral membrane protein LAMP-1 present on the cell surface following stimulation with the nonspecific agonist thioglycollate. Unelicited FIGURE 3. Severe GUSB deficiency is associated with exuberant lyso- peritoneal exudates from naive B6 control mice, as well as somal exocytosis following stimulation in multiple cell types. Naive peritoneal exudate cells from B6, GusbHet,orGusbNull mice were maintained ex GusbNull and GusbHet littermate mice, were collected, and the cells vivo in serum-replacement medium (Control) or in medium with 3% thio- were stimulated ex vivo for 2 h and analyzed by flow cytometry. glycollate (+ 3% Thio) for 2 h and then washed and analyzed by flow We found that unstimulated peritoneal exudate cells from these cytometry. Cell surface LAMP-1 positivity was measured by the percentage strains exhibited a uniformly low LAMP-1 baseline condition of LAMP-1–positive cells (A) and by the mean fluorescence intensity of all (Fig. 3A, 3B, Control). Thioglycollate treatment of naive perito- single cells (B). Significance assessed by one-way ANOVA followed by neal cells markedly induced LAMP-1 cell surface staining in all Bonferroni multiple comparison test (pooled data from three independent groups, as reflected in an increased percentage of LAMP-1–pos- experiments, n = 3 to 9 samples/group; overall p value , 0.0001). (C) itive cells and in the elevated mean fluorescence intensity of all Overlay of representative flow cytometry plots of stimulated (3% Thio, blue) Het GusbNull and unstimulated (Medium, red) peritoneal exudate obtained from Gusb single cells (Fig. 3A, 3B, +3% Thio). cells exhibited Null significantly greater LAMP-1 cell surface staining than either B6 (left panel)orGusb (right panel) mice. Numbers at each corner of the flow cytometry plots represent the percentage of stimulated cells located in or GusbHet cells by each of these metrics (p , 0.001). each quadrant. The LAMP-1 horizontal axes for all four plots are on an To determine which exudate cell types were responsible for this equivalent scale. Quadrants for F4/80-positive peritoneal macrophages (top elevated LAMP-1 positivity, the experiments were repeated using panel) are drawn to separate LAMP-1 Low from LAMP-1 High cells. + a mixture of additional cell type–specific Abs. CD19 B cells (60– Quadrants for CD19-positive peritoneal B-bells (bottom panel) are drawn to + 80%) and F4/80 macrophages (10–25%) comprise the largest separate LAMP-1–negative from LAMP-1–positive cells. ***p , 0.001, percentage of cells in the unelicited peritoneal exudate of naive ****p , 0.0001. The Journal of Immunology 1651 Downloaded from

FIGURE 4. GUSB hypomorphism is associated with exuberant lysosomal exocytosis following stimulation in multiple cell types. BMDM from B6, B6- Gusbh, and C3H strains were cultured for 24 h in the absence or presence of either a low (7.4 3 105/ml) or high (7.4 3 106/ml) concentration of live B. burgdorferi. Lysosomal exocytosis was evaluated by testing cell-free supernatants (A) and cell extracts (B) for BGAL activity (pooled triplicate samples http://www.jimmunol.org/ from three independent experiments, n = 9 samples/group; overall p value , 0.001). Notably, exocytosis is signiﬁcantly induced by the low concentration in C3H BMDM (p , 0.0001, Student t test), but not in the two strains on a B6 genetic background. Cell-free supernatants (C) and cell extracts (D) from MEFs were evaluated in a similar manner using high B. burgdorferi concentration (pooled triplicate samples from two independent experiments, n = 6 samples/ group; overall p value , 0.0001). Signiﬁcance assessed by one-way ANOVA followed by Bonferroni multiple comparison test of all B6 and B6-Gusbh samples. *p , 0.05, **p , 0.01, ***p , 0.001, ****p , 0.0001. CTL, control. mice (data not shown). Under basal conditions, naive peritoneal stimulation, a subset of GusbHet peritoneal macrophages further macrophages from both GusbHet and GusbNull mice are LAMP-1 mobilized LAMP-1 onto the cell surface to adopt a LAMP-1 high Null positive, whereas peritoneal B cells from both strains are pre- phenotype. In contrast, Gusb peritoneal macrophages almost by guest on October 5, 2021 dominantly LAMP-1 negative (Fig. 3C). Following thioglycollate exclusively adopted a LAMP-1 high phenotype following stimu-

FIGURE 5. GUSB hypomorphism is associated with elevated MMP-9 release following stimulation. (A) Gelatin zymography for MMP-9 enzyme activity, using cell-free supernatants from B6, B6-Gusbh, and C3H BMDM, with or without coincubation with live B. burgdorferi (+ B.b.). Top panel shows two adjacent zymography gels representative of three independent experiments; bottom panel shows the Image J (National Institutes of Health) band densitometry traces used for quantitation. (B) Statistical comparison of gelatin zymography data (pooled triplicate samples from three independent experiments, n = 9 samples/group; overall p value , 0.0001). (C) MMP-9 protein levels of cell-free BMDM supernatants, evaluated by ELISA (pooled triplicate samples from two independent experiments, n = 6 samples/group; overall p value , 0.0001). Significance assessed by one-way ANOVA followed by Bonferroni multiple comparison test of all B6 and B6-Gusbh samples. *p , 0.05, ***p , 0.001, ****p , 0.0001. 1652 Gusb MODULATES LYSOSOMAL TRAFFICKING AND MMP-9 SECRETION lation, with very few LAMP-1 low cells observed. Stimulated supernatants were significantly elevated above those of B6 (Fig. 5B, GusbHet peritoneal B cells exhibited minimal LAMP-1 positivity, p , 0.001). In this assay, C3H-derived BMDM supernatants dis- but strikingly, a majority of stimulated GusbNull peritoneal B cells played an elevated basal level of MMP-9 activity, which was further became LAMP-1 positive. These data indicate that severe GUSB induced by B. burgdorferi coincubation, suggesting that other un- deficiency impacts lysosomal trafficking and fusion with the plasma known genetic differences in this strain influence MMP-9 levels. To membrane in response to cellular activation in multiple cell types. validate this result with another experimental approach, supernatant Based on previous reports suggesting that B. burgdorferi coincu- MMP-9 protein levels were also evaluated by ELISA (Fig. 5C). bation induces exocytosis in a variety of myeloid cell types (12), Both the elevated basal MMP-9 levels in C3H-derived cells and the a second experimental approach was pursued to evaluate the impact exuberant release of MMP-9 associated with the Gusbh allele were of Gusb hypomorphism on lysosomal exocytosis. BMDM were corroborated by this second approach. generated from B6, Gusb hypomorphic (B6-Gusbh), and C3H mice. Based on a previous report that rapamycin treatment suppresses Cells were plated overnight, washed, and stimulated by coincubation lysosome exocytosis (15), we examined the impact of rapamycin with live B. burgdorferi in serum-replacement medium. After 24 h of treatment on BGAL release in vitro. MEFs derived from B6-Gusbh coincubation, cell-free supernatants were harvested and analyzed for mice were stimulated with 5 mg/ml sonicated B. burgdorferi for release of the lysosomal enzyme BGAL using a fluorogenic enzy- 24 h in serum-replacement medium containing either vehicle or matic activity assay. We observed that coincubation with a high 100 nmol rapamycin. We found that rapamycin significantly re- concentration of B. burgdorferi was associated with a marked induced release of BGAL into supernatants (Fig. 6). To determine if duction of BGAL release into the supernatant by macrophages from transcriptional downregulation could be responsible, transcripts all three strains (Fig. 4A, HIGH). Interestingly, despite an otherwise from the BGAL gene (Glb1) were assessed by quantitative RT- shared B6 genetic background, BGAL release by B6-Gusbh macro- PCR (qRT-PCR). We found that rapamycin caused no significant Downloaded from phages following coincubation was significantly elevated above that change in the abundance of BGAL transcripts relative to vehicle- of B6 control macrophages. In addition, BGAL activity in corre- treated control samples in MEFs (data not shown). sponding cell extracts was constant across the two strains under basal Next, we evaluated the impact of daily rapamycin treatment on or stimulated conditions (Fig. 4B), suggesting that the observed in- arthritis severity following K/B3N treatment. We observed that crease in supernatant activity is not a consequence of bulk changes in K/B3N-treated B6-Gusbh congenic mice receiving rapamycin de- cellular BGAL production. Coincubation with a 10-fold lower con- veloped significantly less severe ankle swelling than those treated http://www.jimmunol.org/ centration of live B. burgdorferi (Fig. 4A, LOW) triggered consid- with vehicle (Fig. 7A, p , 0.01). Rapamycin treatment was associ- erable BGAL release by BMDM from C3H mice, but not by BMDM ated with reduced serum BGAL levels, consistent with inhibition of from either B6 strain. This confirms the dose dependency of the effect bulk cellular release (Fig. 7B). qRT-PCR analyses of K/B3N-treated and indicates that other unidentified genetic differences between B6 mice were conducted at day 7 on spleens as a tissue distant from the and C3H strains also influence this cellular response. inflammatory response and on ankle joints as the primary site of Due to the intriguing effect of severe Gusb deficiency on lyso- inflammation, and we found that our daily systemic delivery of somal trafficking in nonmyeloid cells (Fig. 3C), we next investi- rapamycin had no significant impact on the abundance of BGAL gated whether Gusb hypomorphism would have a similar impact on transcripts in either case (Fig. 7C, 7D). These findings are consistent by guest on October 5, 2021 fibroblasts, a nonmyeloid cell type that is potentially relevant to with our results in MEFs, in which rapamycin mitigated cellular disease pathogenesis in Lyme and rheumatoid arthritis. We cultured release of BGAL without altering its transcription. Alcian blue mouse embryonic fibroblasts (MEFs) derived from B6, B6-Gusbh, staining was also used to visualize the marked differences in ankle and C3H strains and measured their levels of BGAL release fol- joint histology between mice treated with rapamycin or vehicle, as lowing 24 h of coincubation with live B. burgdorferi.B6control indicated (Fig. 7E, 7F). MEFs displayed no significant change in BGAL release following Then we evaluated rapamycin on Lyme arthritis severity fol- stimulation, whereas B6-Gusbh MEFs exhibited a significantly el- lowing B. burgdorferi infection. We observed that B6 control mice evated BGAL signal (p , 0.001) (Fig. 4C, 4D). These data suggest developed mild ankle swelling and that vehicle-treated B6-Gusbh that Gusbh allele alone can contribute to elevated lysosomal exocytosis following cell activation in multiple distinct cell types. In- terestingly, the increase observed in B6-Gusbh was not seen in C3H control MEFs, suggesting that the C3H strain may harbor additional unknown balancing genetic polymorphisms expressed in fibroblasts that stabilize their response to cellular activation. Although release of BGAL is a reasonable surrogate marker for lysosomal trafficking, it and other related lysosomal enzymes are catalytically inactive at the near neutral pH of the extracellular space and are therefore unlikely to contribute to pathogenesis in any meaningful way. To investigate a plausible downstream effector, we considered matrix metalloproteinase 9 (MMP-9). MMP-9 is essential for murine arthritogenesis following B. burgdorferi infection (13), and intracellular MMP-9 strongly colocalizes with LAMP-2–positive vesicles by confocal microscopy (14). To determine whether the Gusbh allele leads to elevated MMP-9 release, BMDM supernatants FIGURE 6. Rapamycin reduces lysosomal exocytosis in vitro. MEFs derived from B6-Gusbh mice were cultured for 24 h in the absence or presence of obtained after 24 h of coincubation with a high dose of 7.4 3 106/ml 5 mg/ml sonicated B. burgdorferi. During the same 24 h period, samples were B. burgdorferi were tested for MMP-9 activity by gelatin zymog- h treated with 100 nmol rapamycin or an equivalent volume of vehicle. Statis- raphy (Fig. 5A). Supernatants from unstimulated B6 and B6-Gusb tical analysis of BGAL activity in cell-free supernatants (pooled samples from BMDM exhibited low basal MMP-9 enzymatic activities, which two independent experiments, n = 9 or 12 samples/group; overall p value , were significantly elevated in samples coincubated with live 0.0001). Significance assessed by one-way ANOVA followed by Bonferroni B. burgdorferi. The MMP-9 activities of coincubated B6-Gusbh multiple comparison test. *p , 0.05, ***p , 0.001, ****p , 0.0001. The Journal of Immunology 1653 Downloaded from http://www.jimmunol.org/ by guest on October 5, 2021

FIGURE 7. Rapamycin treatment reduces K/B3N-associated ankle swelling. Total of 100 mlK/B3N serum was administered to B6-Gusbh congenic mice byi.p.injectionondays0and2.Micereceiveddailydosesof8mg/kgrapamycinoranequivalentvolumeofvehicle,beginning2dpriortothefirstserum injection. (A) The influence of rapamycin treatment on K/B3N-associated ankle swelling at day 7 (n = both rear ankles from 8 mice/group). (B) Serum BGAL activity at day 7 of K/B3N-treated mice administered vehicle or rapamycin (n = 8 mice/group). (C and D) BGAL mRNA transcript levels from spleens or ankle joints, respectively. (E and F) Alcian blue–stained histology slides from K/B3N-treated B6-Gusbh mice administered vehicle or rapamycin, as indicated. Total magnification 34. Arrowheads indicate position of the cranial tibial tendon sheath. Significance assessed by unpaired Student t test. **p , 0.01. congenic mice exhibited significantly more severe arthritis although the effective suppression of arthritogenesis conferred by (Fig. 8A, p , 0.05). However, this elevated ankle swelling was Gusb hypomorphism despite elevated bacterial burden is remarkable. significantly reduced in rapamycin-treated B6-Gusbh congenic Taken together, these data are consistent with a cellular model mice (p , 0.01), and this treated group was statistically indis- by which GUSB plays distinct but synergistic roles in GAG tinguishable from our B6 negative controls. Serum BGAL levels homeostasis and vesicular trafficking (Fig. 9). Under resting of B. burgdorferi–infected B6-Gusbh congenic mice were also conditions, GUSB hypomorphism leads to mild intracellular significantly suppressed by rapamycin treatment (Fig. 8B). To accumulation of GAGs. In the absence of a second pathological determine whether rapamycin treatment influenced the bacterial trigger, this accumulation remains asymptomatic, but may prog- burden of infected mice, qRT-PCR analysis was conducted on ress with advancing age (16). Severe GUSB deficiency leads to samples prepared from heart tissue. We found that the bacterial spontaneous intracellular accumulation of large amounts of GAG burden was significantly higher in rapamycin-treated mice (Fig. 8C, in storage vacuoles and induces a spontaneous disease process. p , 0.0001), despite their markedly lower arthritis severity Even among individuals carrying identical deficiency mutations, (Fig. 8A). These data suggest that rapamycin treatment has im- the severity of disease and age of onset may be variable (7). portant and distinct impacts on host defense and arthritogenesis, Challenge with an inflammatory trigger leads to lysosomal fusion, 1654 Gusb MODULATES LYSOSOMAL TRAFFICKING AND MMP-9 SECRETION

cascade. This leads to less severe arthritis in mouse models of both K/B3N-induced and Lyme-associated arthritis, despite the presence of equivalent or even elevated levels of inﬂammatory stimuli.

Discussion Our forward genetic approach identified Gusb as an unexpected regulator of murine arthritogenesis. The novelty of this discovery has refocused the importance of lysosomes and associated intracellular vesicles in the inflammatory response. The homeostatic degradation of GAGs by Gusb and other closely related lysosomal enzymes is an essential biological process that has primarily been studied in the context of rare, severe congenital enzymatic deficiencies that induce a spontaneous disease process. Although the pathological mechanism by which lysosomal storage diseases induce musculoskeletal symptoms is not fully understood, one suggestion is that partially digested substrate fragments may directly activate proinflammatory signaling pathways. This idea is bolstered by numerous studies that have demonstrated the therapeutic benefits of anti-inflammatory drugs in animal models of various lysosomal storage diseases (17, 18). These substrates can be detected in the serum and urine of severely affected Downloaded from individuals, and the presence of large excesses or altered relative proportions of distinct GAGs are used for clinical diagnosis (19). This necessitates a pathway for transport out of cellular lysosomes and storage vacuoles, into the extracellular milieu, and eventually into the systemic circulation. However, we observed that extracellular deposition of GAGs in the joint tissue is minimal under basal conditions, http://www.jimmunol.org/ even in GusbNull mice, despite evidence that intracellular GAG accumulation is proceeding as expected (Figs. 1, 2). Instead, the extracellular deposition of GAGs appears to be linked to a localized inflammatory response and is greatly exacerbated in GusbNull or B6- Gusbh mice. This dichotomy between accumulating excess GAGs intracellularly in the absence of a coincident inflammatory response and depositing GAGs extracellularly during localized pathogenesis strongly infers a role for Gusb in vesicular trafficking. by guest on October 5, 2021 A role for lysosomal trafficking in the pathogenesis of Lyme and rheumatoid arthritis is implied by the elevated levels of lysosomal exoglycosidases, including GUSB and BGAL, in the serum and synovial fluid of patients (20). These levels are reported to be es- pecially high in the synovial fluid of Lyme arthritis patients (21), consistent with a localized response. Alterations in plasma GAG profiles have also been identified in patients with rheumatoid arthritis and correlated with the level of ongoing disease activity (22). FIGURE 8. Rapamycin treatment reduces Lyme-associated ankle swelling. (A) Lyme-associated ankle swelling measurements of B6-Gusbh mice Vesicular trafficking is vital to the regulation of cellular in- administered vehicle or 8 mg/kg rapamycin daily by i.p. injection, and wild- flammatory responses, influencing the expression of adhesion type B6 and C3H controls (n =bothrearanklesfrom4to5mice/group; molecules and the release of cytokines, chemokines, ATP, and other overall p =0.0002).(B) Influence of rapamycin treatment on serum BGAL effector molecules from a variety of specialized granules, secretory activity detected at 4 wk of infection (n = 4 to 5 mice/group). (C) B. burg- lysosomes, and related vesicles (23, 24). We observed that in- dorferi burden of hearts isolated from B6-Gusbh mice administered vehicle flammatory stimuli induce lysosomal exocytosis in peritoneal or rapamycin, and wild-type B6 and C3H controls (n =4to5mice/group). B cells, macrophages, and fibroblasts (Figs. 3, 4), and this cellular Significance assessed by one-way ANOVA followed by Bonferroni multiple process may be uniformly conserved (25). Lysosomes are a vital A C comparison test of all B6 background mice ( and ) or by unpaired Student source of lipids during plasma membrane repair (26). This repair t test (B). *p , 0.05, **p , 0.01, ****p , 0.0001. AU, arbitrary units. process is hijacked by the protozoan Trypanosoma cruzi to gain entry to the cell (27) and is essential for cellular retention of the cell surface localization of LAMP-1, and exocytosis of MMP-9 parasite (28). Remarkably, C3H inbred mice are highly susceptible and other lysosomal components. In GUSB hypomorphic or de- to T. cruzi infection, whereas B6 mice are resistant, and this C3H ficient cells, the magnitude of lysosomal exocytosis is elevated, susceptibility correlates with exacerbated macrophage activation and this is accompanied by release of intracellular GAGs. These and cytokine release in vitro (29). extracellular GAGs are then deposited within the extracellular In addition to Gusb, severe deficiencies in 10 other GAG- space, and colocalize with areas of histopathology. Together, these degradative lysosomal enzymes (Idua, Ids, Gns, Hgsnat, Naglu, distinct effects of GUSB insufficiency on trafficking and GAG Sgsh, Galns, Glb1, Arsb, and Hyal1) also result in mucopoly- homeostasis shift the normally protective inflammatory response saccharidosis (MPS I–IX, collectively), and many exhibit muscu- into a pathological cascade. Rapamycin treatment is effective in loskeletal manifestations (30). Severe deficiencies in an additional inhibiting this inflammatory exocytosis, reducing extracellular and seven genes (Asah1, Ctsa, Gba, Gnptab, Gnptg, Man2b1,andNeu1) serum exoglycosidase levels, and suppressing this pathological linked to lysosomal storage disease of other substrates also produce The Journal of Immunology 1655 Downloaded from http://www.jimmunol.org/ by guest on October 5, 2021

FIGURE 9. Model of alterations in lysosomal composition and trafficking related to Gusb genotype. Top panels depict homeostatic changes in lysosomal composition determined by lysosomal GUSB levels. Bottom panels depict alterations in cellular response following inflammatory stimulation. In resting cells, LAMP-1 is predominantly localized to lysosomal membranes, with GUSB, BGAL, and MMP-9 proteins sequestered within intracellular vesicles. Resting Gusb hypomorphic cells develop mild lysosomal GAG storage, with spontaneous formation of enlarged GAG storage vesicles in GusbNull-deficient cells. Inflammatory stimuli induce lysosomal fusion and release of lysosomal contents in all groups. Gusb hypomorphism or deficiency each increase the amount of lysosomal fusion following stimulation, as detected by measuring BGAL and MMP-9 release or LAMP-1 cell surface staining, respectively. These alterations in lysosomal trafficking and composition contribute to downstream inflammatory responses, culminating in altered Lyme and rheumatoid arthritis severity (lightning bolts).WT,wild-type. joint/skeletal manifestations (31). It is noteworthy that among this hibitor of lysosomal exocytosis (15). We corroborated these latter group, neuraminidase (Neu1) has previously been shown to findings in vitro and in vivo, observing marked reductions of regulate lysosomal exocytosis (32), suggesting that perhaps other BGAL abundance in cell-free supernatants and serum following lysosomal enzymes may have a similar effect. rapamycin treatment. mTOR has been proposed as a potential The potentiated release of MMP-9 we observed in Gusbh cells therapeutic target in clinical osteoarthritis, and this is supported by following stimulation with live B. burgdorferi provides a direct a correlation between mTOR expression and disease in patients link to a known effector of Lyme arthritogenesis (Fig. 5), but also (36) and by preclinical mouse models (37). Introduction of rapa- highlights the wide impact that alterations in lysosome exocytosis mycin was also reported to induce long-term disease remission in could have on a variety of proinflammatory effectors. In particular, a patient with juvenile rheumatoid arthritis (38). An earlier study the extracellular release of intracellular GAGs may trigger local- found that rapamycin blunted the development of adjuvant ar- ized TLR activation (18), modulate cytokine signaling (33, 34), or thritis in rats (39). We found that rapamycin effectively suppressed contribute to the establishment of chemokine gradients (35). the elevated arthritis severity caused by Gusb hypomorphism in These effects are not mutually exclusive and could all act in our mouse models of K/B3N and Lyme arthritis (Figs. 7, 8). concert upon arrival of infectious Borrelia in the joint to establish Because K/B3N arthritis represents the effector phase of disease a proarthritogenic state (Fig. 9). progression, and because Lyme arthritis severity was reduced The mammalian target of rapamycin (mTOR) inhibitor rapa- despite a marked increase in the bacterial burden of rapamycin mycin has been previously described as a pharmacological in- treated animals, this infers that mTOR activation plays an im- 1656 Gusb MODULATES LYSOSOMAL TRAFFICKING AND MMP-9 SECRETION portant role in the pathogenic cascade induced by the host re- 16. Yatziv, S., R. P. Erickson, R. Sandman, and W. V. Robertson. 1978. Glycos- aminoglycan accumulation with partial deficiency of beta-glucuronidase in the sponse, downstream of the initial inflammatory triggers. C3H strain of mice. Biochem. Genet. 16: 1079–1084. This research has clear implications for human populations, in 17. Eliyahu, E., T. Wolfson, Y. Ge, K. J. Jepsen, E. H. Schuchman, and whom GUSB levels are widely variable (40, 41). Even more in- C. M. Simonaro. 2011. Anti-TNF-alpha therapy enhances the effects of enzyme replacement therapy in rats with mucopolysaccharidosis type VI. PLoS One 6: triguing, the reported elevation of circulating lysosomal enzymes e22447. and alterations in GAG profiles within general populations of 18. Simonaro, C. M., Y. Ge, E. Eliyahu, X. He, K. J. Jepsen, and E. H. Schuchman. patients with arthritis suggest that lysosomal exocytosis may be 2010. Involvement of the Toll-like receptor 4 pathway and use of TNF-alpha antagonists for treatment of the mucopolysaccharidoses. Proc. Natl. Acad. Sci. a conserved process during inflammatory arthritogenesis, raising USA 107: 222–227. its potential diagnostic and therapeutic value. The large impact 19. Lehman, T. J., N. Miller, B. Norquist, L. Underhill, and J. Keutzer. 2011. Di- that GUSB hypomorphism has on experimental arthritis severity agnosis of the mucopolysaccharidoses. Rheumatology (Oxford) 50(Suppl 5): v41–v48. and its ubiquitous expression in all cell types infer that lysosomal 20. Pancewicz, S. A., P. Wielgat, T. Hermanowska-Szpakowicz, M. Kondrusik, function and trafficking are important determinants of the patho- J. Zajkowska, S. Grygorczuk, J. Popko, and K. Zwierz. 2006. Activity of lysosomal exoglycosidases in the serum of patients with chronic Lyme arthritis. Int. logical trajectory of inflammation following a primary insult. J. Med. Microbiol. 296(Suppl 40): 280–282. 21. Pancewicz, S., J. Popko, R. Rutkowski, M. Knas, S. Grygorczuk, T. Guszczyn, M. Bruczko, S. Szajda, J. Zajkowska, M. Kondrusik, et al. 2009. Activity of Acknowledgments lysosomal exoglycosidases in serum and synovial fluid in patients with chronic We thank Robert Lochhead and Lynn Sonderegger for providing helpful Lyme and rheumatoid arthritis. Scand. J. Infect. Dis. 41: 584–589. discussions and suggestions; Katie Bashant, Ky-phuong Luong, and Anne 22. Jura-Po´łtorak, A., K. Komosinska-Vassev, A. Kotulska, E. J. Kucharz, Brogdon for sample preparation and technical assistance; and Jackie K. Klimek, M. Kopec-Medrek, and K. Olczyk. 2014. Alterations of plasma glycosaminoglycan profile in patients with rheumatoid arthritis in relation to Paquette and Sarah Whiteside for reviewing the manuscript. Electron disease activity. Clin. Chim. Acta 433: 20–27. Downloaded from microscopy was performed by Nancy Chandler at the University of Utah 23. Appelqvist, H., P. Wa¨ster, K. Ka˚gedal, and K. O¨ llinger. 2013. The lysosome: Electron Microscopy Core Laboratory. from waste bag to potential therapeutic target. J. Mol. Cell Biol. 5: 214–226. 24. Wernersson, S., and G. Pejler. 2014. Mast cell secretory granules: armed for battle. Nat. Rev. Immunol. 14: 478–494. Disclosures 25. Rodrı´guez, A., P. Webster, J. Ortego, and N. W. Andrews. 1997. Lysosomes The authors have no financial conflicts of interest. behave as Ca2+-regulated exocytic vesicles in fibroblasts and epithelial cells. J. Cell Biol. 137: 93–104. 26. Andrews, N. W., P. E. Almeida, and M. Corrotte. 2014. Damage control: cellular

mechanisms of plasma membrane repair. Trends Cell Biol. 24: 734–742. http://www.jimmunol.org/ References 27. Tardieux, I., P. Webster, J. Ravesloot, W. Boron, J. A. Lunn, J. E. Heuser, and 1. Kuehn, B. M. 2013. CDC estimates 300,000 US cases of Lyme disease annually. N. W. Andrews. 1992. Lysosome recruitment and fusion are early events re- JAMA 310: 1110. quired for trypanosome invasion of mammalian cells. Cell 71: 1117–1130. 2. Radolf, J. D., M. J. Caimano, B. Stevenson, and L. T. Hu. 2012. Of ticks, mice 28. Andrade, L. O., and N. W. Andrews. 2004. Lysosomal fusion is essential for the and men: understanding the dual-host lifestyle of Lyme disease spirochaetes. retention of Trypanosoma cruzi inside host cells. J. Exp. Med. 200: 1135–1143. Nat. Rev. Microbiol. 10: 87–99. 29. Russo, M., N. Starobinas, R. Ribeiro-Dos-Santos, P. Minoprio, H. Eisen, and 3. Steere, A. C., R. T. Schoen, and E. Taylor. 1987. The clinical evolution of Lyme M. Hontebeyrie-Joskowicz. 1989. Susceptible mice present higher macrophage activation than resistant mice during infections with myotropic strains of Try- arthritis. Ann. Intern. Med. 107: 725–731. panosoma cruzi. Parasite Immunol. 11: 385–395. 4. Shapiro, E. D., R. Dattwyler, R. B. Nadelman, and G. P. Wormser. 2005. Re- 30. Morishita, K., and R. E. Petty. 2011. Musculoskeletal manifestations of muco- sponse to meta-analysis of Lyme borreliosis symptoms. Int. J. Epidemiol. 34: polysaccharidoses. Rheumatology (Oxford) 50(Suppl 5): v19–v25. 1437–1439; author reply 1440–1433. 31. Aldenhoven, M., R. J. Sakkers, J. Boelens, T. J. de Koning, and N. M. Wulffraat. by guest on October 5, 2021 5. Barthold, S. W., D. S. Beck, G. M. Hansen, G. A. Terwilliger, and K. D. Moody. 2009. Musculoskeletal manifestations of lysosomal storage disorders. Ann. 1990. Lyme borreliosis in selected strains and ages of laboratory mice. J. Infect. Rheum. Dis. 68: 1659–1665. Dis. 162: 133–138. 32. Yogalingam, G., E. J. Bonten, D. van de Vlekkert, H. Hu, S. Moshiach, 6. Bramwell, K. K., Y. Ma, J. H. Weis, X. Chen, J. F. Zachary, C. Teuscher, and S. A. Connell, and A. d’Azzo. 2008. Neuraminidase 1 is a negative regulator of J. J. Weis. 2014. Lysosomal b-glucuronidase regulates Lyme and rheumatoid lysosomal exocytosis. Dev. Cell 15: 74–86. arthritis severity. J. Clin. Invest. 124: 311–320. 33. Yoshioka, Y., E. Kozawa, H. Urakawa, E. Arai, N. Futamura, L. Zhuo, 7. Tomatsu, S., A. M. Montanõ, V. C. Dung, J. H. Grubb, and W. S. Sly. 2009. K. Kimata, N. Ishiguro, and Y. Nishida. 2013. Suppression of hyaluronan syn- Mutations and polymorphisms in GUSB gene in mucopolysaccharidosis VII (Sly thesis alleviates inflammatory responses in murine arthritis and in human Syndrome). Hum. Mutat. 30: 511–519. rheumatoid synovial fibroblasts. Arthritis Rheum. 65: 1160–1170. 8. Ray, A., and B. N. Dittel. 2010. Isolation of mouse peritoneal cavity cells. J. Vis. 34. Petrey, A. C., and C. A. de la Motte. 2014. Hyaluronan, a crucial regulator of Exp. 35: 1488. inflammation. Front. Immunol. 5: 101. 9. Ma, Y., K. P. Seiler, K. F. Tai, L. Yang, M. Woods, and J. J. Weis. 1994. Outer 35. Johnson, Z., A. E. Proudfoot, and T. M. Handel. 2005. Interaction of chemokines surface lipoproteins of Borrelia burgdorferi stimulate nitric oxide production by and glycosaminoglycans: a new twist in the regulation of chemokine function the cytokine-inducible pathway. Infect. Immun. 62: 3663–3671. with opportunities for therapeutic intervention. Cytokine Growth Factor Rev. 16: 10. Bramwell, K. K. C., C. Teuscher, and J. J. Weis. 2014. Forward genetic 625–636. approaches for elucidation of novel regulators of Lyme arthritis severity. Front. 36. Tchetina, E. V., A. R. Poole, E. M. Zaitseva, E. P. Sharapova, N. G. Kashevarova, Cell. Infect. Microbiol. 4: 76. E. A. Taskina, L. I. Alekseeva, L. A. Semyonova, S. I. Glukhova, A. N. Kuzin, 11. Vogler, C., B. Levy, N. Galvin, M. S. Sands, E. H. Birkenmeier, W. S. Sly, and et al. 2013. Differences in Mammalian target of rapamycin gene expression in J. Barker. 2001. A novel model of murine mucopolysaccharidosis type VII due to the peripheral blood and articular cartilages of osteoarthritic patients and disease an intracisternal a particle element transposition into the beta-glucuronidase activity. Arthritis (Egypt) 2013: 461486. 37. Pal, B., H. Endisha, Y. Zhang, and M. Kapoor. 2015. mTOR: a potential ther- gene: clinical and pathologic findings. Pediatr. Res. 49: 342–348. apeutic target in osteoarthritis? Drugs R D. 15: 27–36. 12. Gebbia, J. A., J. L. Coleman, and J. L. Benach. 2001. Borrelia spirochetes up- 38. Foroncewicz, B., K. Mucha, L. Paczek, A. Chmura, and W. Rowinski. 2005. regulate release and activation of matrix metalloproteinase gelatinase B (MMP- Efficacy of rapamycin in patient with juvenile rheumatoid arthritis. Transpl. Int. Infect. Immun. 9) and collagenase 1 (MMP-1) in human cells. 69: 456–462. 18: 366–368. 13. Heilpern, A. J., W. Wertheim, J. He, G. Perides, R. T. Bronson, and L. T. Hu. 39. Carlson, R. P., D. A. Hartman, L. A. Tomchek, T. L. Walter, J. R. Lugay, 2009. Matrix metalloproteinase 9 plays a key role in lyme arthritis but not in W. Calhoun, S. N. Sehgal, and J. Y. Chang. 1993. Rapamycin, a potential dissemination of Borrelia burgdorferi. Infect. Immun. 77: 2643–2649. disease-modifying antiarthritic drug. J. Pharmacol. Exp. Ther. 266: 1125–1138. 14. Sbai, O., A. Ould-Yahoui, L. Ferhat, Y. Gueye, A. Bernard, E. Charrat, 40. Lombardo, A., C. Bairati, G. Goi, C. Roggi, L. Maccarini, D. Bollini, and A. Mehanna, J. J. Risso, J. P. Chauvin, E. Fenouillet, et al. 2010. Differential A. Burlina. 1996. Plasma lysosomal glycohydrolases in a general population. vesicular distribution and trafficking of MMP-2, MMP-9, and their inhibitors in Clin. Chim. Acta 247: 39–49. astrocytes. Glia 58: 344–366. 41. Sperker, B., T. E. Murdter,€ M. Schick, K. Eckhardt, K. Bosslet, and 15. Martins, R. M., R. M. Alves, S. Macedo, and N. Yoshida. 2011. Starvation and H. K. Kroemer. 1997. Interindividual variability in expression and activity of rapamycin differentially regulate host cell lysosome exocytosis and invasion by human beta-glucuronidase in liver and kidney: consequences for drug metabo- Trypanosoma cruzi metacyclic forms. Cell. Microbiol. 13: 943–954. lism. J. Pharmacol. Exp. Ther. 281: 914–920.