Fig 2: LRG1 modifies the effects of TGFß on myeloid and hematopoietic progenitor cell growth.(A) HL-60 cell proliferation in the presence of increasing concentrations of TGFß alone (blue diamonds) or TGFß plus LRG1 (800 ng/mL, red squares) was assessed using the XTT assay. A dose-response reduction in cell proliferation as measured by the decreasing A490-A650 is observed with increasing concentrations of TGFß that is mitigated by the addition of LRG1. The Student's t-test was used to determine the significance of the differences between control and LRG1 treated samples († p < 0.05) (n = 5). (B) Purified bone marrow-derived CD34+ cells were cultured in serum-free semi-solid media in the presence of TGFß (10 ng/mL) with or without LRG1 at the indicated concentrations, and colony numbers scored. An increase in colony numbers relative to TGFß1 treatment alone (0) is seen with increasing concentrations of LRG1. The Student's t-test was used to determine the significance of the differences between control and LRG1 treated samples († p < 0.05) (n = 3). (C) The effects of TGFß alone (left panel) and TGFß plus LRG1 (right panel) on colony formation and colony composition were analyzed. Upper two images in each panel. Representative light microscopy images of methocult plates on which bone marrow-derived CD34+ cells were cultured in the presence of TGFß alone (left) or TGFß with LRG1 (right). Lower two images in each panel. Wright-Giemsa staining of cells plucked from individual colonies from the methocult plates in the upper images is shown. Cells were grown as described in (B) in the presence of TGFß (10 ng/mL) with or without LRG1 (0.625 ug/mL). LRG1 mitigates the inhibitory effect of TGFß on the growth and size of colony-forming units-granulocyte monocyte (CFU-GM). The data shown are representative of five independent experiments, each with a different normal donor.

Fig 3: Hierarchical clustering of T1D and HS samples using the expression profiles of 30 differentially regulated proteins (see Table 2). The x-axis displays protein names while the y-axis displays clusters of the 210 subjects. The bar on the right denotes subjects in a two-color code (green for healthy subjects; red for T1D patients) and a three-color code based on HbA1c values (green for healthy subjects; red for medium levels and purple for high levels of HbA1c). Protein abundances (log values) are displayed in a red-to-blue color code (high to low abundance). The proteins COL12A1, NAGA, FGFR2, LRG1, A1BG and CD14 drive the separation of the two main clusters.

Fig 4: Section of protein-protein interaction network including the interactions of 29 proteins with differential abundance in the T1D and HS cohorts (LRG1, CD14, AZGP1, NAGA, CTSD, CTSB, CPE, and CP). The shade of red indicates the fold change in abundance (light red = > 1.2 and < 1.5; medium red = >1.5 and < 2.0; dark red = >2.0). Only LRG1 has fold change above 2.0 in the above network. A green denotes downregulated proteins with fold change > -1.0 and < -1.5 and include highly connected SDC4 and FN1. A yellow color denotes the absence of statistically significant changes of connected proteins in the network. A grey color denotes protein absence in our proteome datasets. The connection of CTSD and CPE to insulin (INS) pertains to INS protein processing by these peptidases. The separation of subcellular locations in the graphic illustrates the concept of lysosomes as subcellular biomarkers of altered metabolism in individuals with T1D. An insert (top-right) depicts the physical interaction of LRG1 and FN1. Albumin (HSA) served as a negative control of the FN1/LRG1 interaction. The extracellular matrix protein SDC4 did not show an interaction with LRG1 in far-Western blot experiments.

Fig 5: Neutrophil-derived LRG1 is differentially glycosylated but retains affinity for cytochrome c.Neutrophil (PMN) lysates from healthy volunteers were prepared as described. PMN releasates were prepared by exposing isolated neutrophils to calcium ionophore and the resultant supernatants concentrated by ultrafiltration. (A) LRG1 purified from human serum, PMN lysate, and PMN granule releasate were left untreated (- lanes) or subjected to deglycosylation (+ lanes) using a commercially available protein deglycosylation kit, and immunoblot analysis performed using an antibody to human LRG1. Molecular weight markers (MM) are shown on the left. Neutrophil-derived LRG1 migrates with a higher molecular weight than serum-derived LRG1 (- lanes) but with the same molecular weight following deglycosylation (+ lanes), indicating different patterns of glycosylation of the native protein forms in PMNs compared to serum. The data shown are representative of three independent experiments, each with a different normal donor. (B) Purified cytochrome c was bound to a NHS-activated sepharose column. Releasate from PMNs stimulated with calcium ionophore (2 uM) was loaded onto the cytochrome c column. Bound proteins were eluted with an acetate buffer (pH 4) in 1 mL fractions, and subjected to immunoblot analysis with antibody to LRG1. Lane 1 is crude PMN releasate. The column flow through and wash is shown in lane 2. Eluate fractions are shown in lanes 3–8, with LRG1 detectable in fractions 3 and 4. Lane 9 is purified LRG1. The data shown are representative of three independent experiments.