Faculty Bibliography

INTRODUCTION: Many adult tissues harbor stem cells, which theoretically could be employed for injury repair [1, 2], but the origins of these cells, and the factors that influence their developmental potency, are poorly understood. The skeleton contains tissue-specific stem cells, which are responsible for maintaining bone mass and for regenerating new bone following injury. By genetic cell lineage labeling we established that adult skeletal stem cells come from two embryonic lineages, the mesoderm and the neural crest [3]. Although both populations give rise to cartilage and bone, they are not functionally equivalent: Neural crest (NC)-derived skeletal progenitor cells are more osteogenic, and exhibit robust plasticity in bone grafting assays compared to mesoderm-derived skeletal stem cells [3]. The embryonic origin, however, is not the only attribute that differs between cells from the cranial versus appendicular skeleton. The embryonic Hox code is responsible for positional patterning during development, and our data show that the Hox code still serves as positional memory during adult bone regeneration. The goal of this study was to further elucidate and understand the molecular basis for this remarkable plasticity of the NC-derived, Hox-negative progenitor cell compared to the rather committed mesoderm-derived, Hox-positive progenitor cell. METHODS: The periosteum from four different skeletal sites (F-frontal bone, H-hyoid bone, P-parietal bone, T-tibia) was carefully collected into RNAlater, snap frozen in liquid nitrogen, then RNA isolation was carried out. RNA obtained was inspected on Bioanalyzer and samples with RIN superior to 8 were then subjected to RNA sequencing utilizing high output, paired-end reads using the Illumina HiSeq 2500 System. Three replicates from each skeletal location were tested. Bioinformatic analysis was performed with TopHat (version 2.0.9), Cufflinks (version 2.2.0) and Htseq (version 0.6.1.p.1). The expression of Hox genes with the most relevant expression levels detected by RNAseq was validated by qRT-PCR. RESULTS SECTION: Utilizing RNA sequencing we first set out to test whether the embryonic Hox code continues to be expressed in adult cells, and if the Hox expression pattern matches that of the embryo. We harvested periosteum from four distinct locations, each representing a unique signature of embryonic Hox code (positive/negative) and embryonic origin (neural crest/mesoderm). The periosteum from the tibia (MD, Hox-pos), hyoid (NC, Hox-pos), frontal (NC, Hox-neg) and parietal bone (MD, Hox-neg) was isolated, followed by standard RNA preparation. We then used RNAseq to identify the transcriptome of these four cell origins. First we hypothesized that the embryonic Hox status of periosteal cell is maintained into adulthood. Analysis of the RNAseq data confirmed this hypothesis: Parietal and frontal bones showed insignificant amounts of Hox transcripts, while the tibia and hyoid showed relatively high expression levels (Fig. 1A). Next, we aimed at understanding whether periosteal progenitor cells can be distinguished by their embryonic origin or by their Hox code expression. A first suggestions that the Hox code is a more defining characteristic than the embryonic origin was shown by the hierarchical cluster analysis, in which the progenitors from the tibia and hyoid and the progenitors from the frontal and parietal bone clustered together according to their Hox expression profile (Fig. 1B). We then confirmed this distinction according to the Hox code by plotting the results in an MA plot. Here, every dot represents one gene, genes with an adjusted p value less than 0.01 are shown in red. Comparison of neural crest (frontal and hyoid) and mesoderm (parietal and tibia) derived progenitors revealed a paucity of differentially expressed genes, while comparing Hox-negative (frontal and parietal) and Hox-positive (hyoid and tibia) progenitors resulted in an abundance of differentially expressed genes (Fig. 1C). This analysis suggests that it is the Hox status that characterizes and distinguishes skeletal progenitor cells more accurately than the embryonic origin. DISCUSSION: In previous experiments, we made an unexpected discovery: skeletal stem cells come in at least two "flavors". Using a genetic cell lineage labeling strategy, we identified a mesodermderived population that is responsible for remodeling and repair in long bones, and a second population derived from the NC that remodels and repairs craniofacial bones [3]. These results established for the first time that NC- and mesoderm-derived bones heal through the selective recruitment of skeletal stem/progenitor cells from their own embryonic origins. However, this selective 'flavor' has to somehow be imprinted into the progenitor cell. The Hox code represent a mechanism crucial for embryonic patterning, and therefore we sought to investigate whether the positional memory, defined by the Hox expression pattern, persists into adulthood, and if so, if it represents the mechanism by which adult progenitor cells can be characterized. Our gene transcription analysis confirmed our hypothesis that it is in fact the Hox code that distinguishes progenitor cells more effectively than the embryonic origin. The finding that progenitor cells maintain positional memory throughout their life, and that this identity is associated with a unique transcriptional profile may have significant clinical implications. If bones preferentially heal using cells that share the same positional origin (Hox code), then reparative strategies may have to take this variable into account in order to be maximally effective. (Figure Presented)

Pancreatic cancer is a devastating disease and ranks as the third most common cause of cancer-related death. Like many cancers, there has been increased interest in the role of the immune system in the progression and development of pancreatic cancer. In particular, immunosuppression within the tumor microenvironment (TME) is thought to impair the host's antitumor response. In this article, we review myeloid-derived suppressor cells and their contribution to this immunosuppression within the pancreatic TME.

Since progranulin (PGRN) is a natural ligand of TNF receptors, we assessed whether serum PGRN levels predict and/or reflect responsiveness of RA patients to TNF-antagonist therapy. TNF-antagonist-naive RA patients (N = 35) were started on TNF-antagonist therapy. At baseline and at follow-up visits, DAS28-ESR, DAS28-CRP, and CDAI were calculated, and venous blood was collected for serum PGRN determination. Disease activity and clinical response were based on EULAR criteria. Baseline serum PGRN levels varied considerably and correlated with ESR and CRP. DAS28-ESR, DAS28-CRP, and CDAI were greater in "PGRN-high" than in "PGRN-low". Baseline serum PGRN levels did not predict clinical responsiveness to TNF-antagonist therapy. Nevertheless, changes in serum PGRN levels at 274+ days following initiation of TNF-antagonist therapy correlated with changes in ESR, CRP, DAS28-ESR, DAS28-CRP, and CDAI. At this time, DAS28-ESR, DAS28-CRP, and CDAI in PGRN-high and PGRN-low equalized, but serum PGRN levels remained greater in PGRN-high than in PGRN-low. To our knowledge, the present report is the first prospective study to longitudinally assess changes in serum PGRN levels following initiation of TNF-antagonist therapy. Although pre-treatment serum PGRN levels may not predict clinical responsiveness to TNF-antagonist therapy, changes in serum PGRN levels correlate with changes in disease metrics over time. By inference, administration of PGRN may represent an effective therapeutic option for development in RA patients.

INTRODUCTION: All tissues are affected by aging, but diseases that weaken the skeleton constitute the most prevalent chronic impairment in the United States. Although skeletal diseases and conditions are seldom fatal, they can significantly compromise function and diminish quality of life. Perhaps most importantly, age-related changes in skeletal health can be traced back to a decline in both the number and function of osteoprogenitor cells (OPCs). However, the cause for the decline in both the number and function of OPCs is not well understood. Chronic inflammation in the elderly (inflamm-aging) is thought to be a major contributor to this decline in the regenerative capacity of many tissues, including the skeleton. In contrast to a well-balanced inflammatory response after trauma, which is crucial for successful bone repair, chronic unbalanced elevation of pro-inflammatory cytokines has been shown to inhibit regeneration in a variety of tissues. We hypothesize that inflamm-aging is the major cause for the decline in OPC number and dysfunction in elderly patients and that this decline in OPC number and dysfunction can be halted by treatment with an anti-inflammatory drug. METHODS: Young, 12 week-old and aged, 52 week-old C57BL/6J were used following the IACUC guidelines at our institution. Aged animals were randomly distributed into a no-treatment (n=5) and a treatment group (n=5). Animals in the no-treatment group received regular drinking water, while animals in the treatment group received sodium salicylate water (12mg/day) for 8 weeks. The inflammatory status of young and aged untreated and treated mice was assessed using a multiplex platform screening for multiple pro- and anti-inflammatory cytokines, and utilizing qRT-PCR for IL-1, IL-6, NF-kappaB, TNF-alpha. FACS analysis using the LepR as a marker for osteogenic precursor cells was employed to identify the effect of chronic low-level inflammation on progenitor cell number. In addition, bMSCs were harvested from femurs and tibia from young and aged untreated and treated mice and cultured in growth media and osteogenic media. Cell proliferation and osteogenic differentiation (qRT-PCR for col 1, runx2, osx and oc, and alizarin red and alkaline phosphatase staining) were assessed in vitro. Results are presented in the form of mean +/- standard deviation, with N equal to the number of samples analyzed. Two-tailed Student's t-tests were used to determine significant differences between data sets that are normally distributed. For non-normally distributed data sets, Mann-Whitney U test was used. Significance was attained at p < 0.05 and all statistical analyses were performed with Graphpad Prism software (GraphPad Software, San Diego, California). RESULTS: First, we set out to identify age-related chronic inflammation in mice. We analyzed blood by multiplex analysis and tibial and femoral bone marrow by qRT-PCR for pro-inflammatory markers. Both analyses revealed an increase in pro-inflammatory and a decrease in anti-inflammatory cytokines in aged animals, confirming the presence of inflamm-aging (Fig. 1A). Next, we aimed at understanding how aging effects osteoprogenitor cell number using flow cytometry. We harvested cells from young and aged mice, removed red blood cells, and then stained with antibodies to CD31, CD45, Ter-119 and LepR. Cell sorting was performed and CD31^-CD45^-Ter^-119^-LepR+ cells were isolated and quantified. Flow cytometry analysis revealed that 0.38% of cells from 12 week old mice were LepR⁺ osteoprogenitor cells, confirming the findings published by Zhou et al. Analysis of the cells from 52 week-old mice revealed a significant decrease in number to 0.017% of bone marrow cells (Fig. 1B). Next, we had to establish that an 8-week course of sodium salicylate successfully represses chronic inflammation. In response to NSAID treatment, the expression level of IL-10 significantly increased above the level of the juvenile animals, while the NF-kappaB, TNFalpha and Cox-2 levels returned to baseline (Fig. 1A). This experiment served as a proof-of-principle that mice exhibit inflamm-aging and that this inflammatory state can be suppressed by administration of an NSAID. Having established that our dosing of the systemic NSAID suppresses chronic ageinduced inflammation, we assessed whether OPC frequency changes as a result of suppressed inflamm-aging. Bone marrow from 3 month-old, 12 month-old and 12 monthold NSAID-treated animals was subjected to flow cytometry. OPC frequency declined with aging, however, after an 8 week course of NSAID treatment, we noticed a two-fold increase in LepR⁺ OPCs within the bone marrow (Fig. 1B). We then aimed at testing whether NSAID treatment of aged mice resulted in a restoration of the osteogenic potential of this OPC population. Quantitative RT-PCR of the bone marrow of sodium salicylate-treated mice showed an increase in osteogenic gene expression (osx, oc and alkaline phosphatase) compared to untreated aged mice. The expression level of osx reached that of young, 3 month-old mice, while oc and ALP expression levels were significantly higher than those of juvenile animals (Fig. 1C). In order to further characterize this increase in osteogenic potential, we harvested MSCs from young and aged treated and untreated animals, plated them in vitro and then subjected them to osteogenic differentiation media. Mineralization assays and expression analysis of osx, oc and ALP showed decreased osteogenesis of aged cells, while treatment with sodium salicylate recovered this decline and resulted in restoration of the osteogenic potential (Fig. 1D). DISCUSSION: These experiments demonstrate for the first time that age-related chronic inflammation is responsible for the decreased proliferative and osteogenic potential of aged OPCs and that this process is reversible by anti-inflammatory treatment. The findings from this study may have a profound translational impact: If we could restore the regenerative potential of the aged skeleton by treating age-related inflammation, then theoretically, we may have a tool at hand to improve the healing process of osteoporotic fracture patients. (Figure Presented)