Oxidized La decorates the surface of osteoclasts during multinucleation.

(a) Representative immunofluorescence (top) and differential interference contrast (DIC, bottom) confocal micrographs of permeabilized primary human osteoclasts. La localization was visualized via a general α-La antibody (Pan), an α-La antibody that recognizes oxidized La, or an α-La antibody thar recognizes reduced La (described and validated in {Berndt, 2021 #2}). (b) Representative immunofluorescence confocal micrographs of primary human osteoclasts stained with the antibodies described in a under non-permeabilized conditions to visualize surface La. Surface La pools were visualized for the untreated cells (control) and for cells treated with the membrane impermeable reducing reagent TCAP. (c) Quantification of b. (n=3) (p= 0.13, 0.0001, and 0.03, respectively. Statistical significance evaluated via paired t-test.

Oxidized La promotes osteoclast membrane fusion

(a) Representative fluorescence and DIC confocal micrographs of primary human osteoclasts following synchronized cell-cell fusion in control conditions or in the presence of α-La antibodies. Cyan=Hoechst (b) Quantification of a. (n=5) (p= 0.0001, 0.001, 0.1, and 0.0008, respectively) (c) Quantification of synchronized primary human osteoclast fusion events under control conditions or conditions where surface proteins are oxidized (TCEP). Osteoclast fusion was synchronized by reversibly inhibiting cell-cell fusion using the membrane fusion inhibitor LPC. Inhibition=LPC applied and not removed, Wash=LPC applied and removed to allow synchronized fusion, TCEP=same as Wash with the addition of TCEP before LPC removal. (n=4, except 25 where n=2) (p= 0.005, 0.054, 0.04, 0.02, and 0.005, respectively) Statistical significance evaluated via paired t-test.

Surface La’s oxidation is functionally important for the formation and function of osteoclasts.

(a) Representative fluorescence images of primary human osteoclast multinucleation following addition of control La 194-408 vs La 194-408 where cysteine residues were reduced by TCEP and then blocked by iodoacetamide treatments. Cyan=Hoechst. Magenta=Phalloidin (b) Quantification of osteoclast fusion events in a. (n=3) (p= 0.008 and 0.04, respectively) (c) Quantification of in vitro resorptive function in conditions described in a. (n=4) (p= 0.007 and 0.009, respectively) Statistical significance evaluated via paired t-test.

La 194-408 cysteine residues are critical for promoting osteoclast fusion.

(a) Cartoons illustrating the domain structure of La’s C-terminal half and the location of its two cysteines. (b) Representative Western Blots depicting La C-terminal half and cystine mutant. (c) Representative immunofluorescence micrographs of fusing primary human osteoclasts under control conditions or treated with La 194-408 or cysteine mutant La 194-408. (d) Quantification of the number of fusion events observed in c. (n=4) (p= 0.004, 0.02, and 0.26, respectively) Statistical significance evaluated via paired t-test.

ROS promotes oxidized La surface trafficking and osteoclast fusion.

(a) Representative confocal micrographs of ROS signal in primary human osteoclasts precursors under conditions lacking RANKL, following 16 hours of RANKL application, or following 16 hours and 1hr 100 μM NAC treatment. (b) Quantification of ROS signaling in osteoclast progenitors, committed osteoclasts, or committed osteoclasts treated with the membrane permeable reducing reagent NAC (n=2, 4, and 3, respectively) (p= 0.03 and 0.01, respectively) (b) qPCR quantification of osteoclastogenesis markers (NFATc1, cFOS), La transcript (SSB), and annexin A5 (ANXA5). Expression evaluated in comparison to GAPDH. (n=4) (p= 0.50, 0.69, 0.32, and 0.45, respectively) (c) Quantification of fusion events at day 3 post RANKL application between primary human osteoclasts under control conditions or in the presence of NAC from day 1 post RANKL application. (n=6) (p= 0.009, 0.0005, respectfully) (d) Quantification of La surface staining of non-permeabilized cells with pan a-La antibodies at Day 3 post RANKL application without or with 50-100 μM NAC added at day 1 post RANKL application. (n=9) (p= <0.0001) (e) Quantification of the number of fusion events observed between human osteoclasts in control conditions, conditions where fusion was inhibited via NAC treatment as in d, and where fusion was inhibited via NAC and rescued with La 194-408 or cysteine mutant La 194-408. (n=5) (p= 0.0007, 0.0001, .0054, and 0.0352, respectfully). Statistical significance evaluated via paired t-test.

ROS signaling promotes dephosphorylation of La and osteoclast fusion by increasing the amounts of oxidized La at the surface of the cells.

(a) Fluorescence microscopy and DIC images of permeabilized osteoclasts without or with application of 100 μm NAC (1h) stained with an α-La antibody that recognizes La phosphorylated at Ser366. (b) Quantification of the staining intensity from a. (n=3) (p=0.01). (c) Representative fluorescence images of differentiating osteoclasts in control conditions, conditions where fusion was inhibited via NAC treatment (100 μM NAC added at day 2 days post RANKL application), and conditions where fusion was rescued by the application of recombinant La 194-408 or cysteine mutant La 194-408. (d) Quantification of the number of osteoclast fusion events in c. (n=5) (p= 0.0007, 0.0001, 0.0054, and 0.0352, respectfully). Statistical significance evaluated via paired t-test.

ROS signaling induced restructuring of La from reduced to oxidized species triggers La re-localization from nucleus to the surface of differentiating osteoclasts promoting their fusion and resorptive function.

Osteoclastogenic differentiation from monocytes to mature bone-resorbing osteoclasts initiated by subsequent application of M-CSF and RANKL depends on intracellular ROS induced drastic changes in the redox state and localization of La: from nuclear, reduced species of La in monocytes and macrophages to cell-surface associated, oxidized dephosphorylated species La in fusion-competent osteoclasts and, when fusion stops, back to nuclear. reduced species of La in mature multinucleated osteoclasts.

Oxidized La is found in the cytosol.

3D stack image depicting the cytosolic localization of oxidized LA. Top - XY-slice slightly below the equatorial plane of a permeabilized multinucleated osteoclast (3 days post-RANKL application) stained with an α-La antibody that recognizes oxidized La. 3D stack of this representative cell was acquired with 0.22μm step interval. Bottom - Z-slice through the orange line on the top panel. Red ellipses show the approximate outlines of two nuclei in the slice, and the orange line shows the location of the XY-slice shown on top.

La C-terminal half and cysteine mutant purification.

(a) A gray-scale image of La 194-408 or cysteine mutant La 194-408 separated via polyacrylamide gel electrophoresis and visualized using Coomassie staining. Representative Western Blots depicting La C-terminal half and cysteine mutant recognized by α-6xhis (b) or α-La (ox.) α-6xhis (c). < Denotes the migration of La 194-408 as a dimer.

NAC inhibits redox shift from reduced to oxidized species of intracellular La.

(a) Fluorescence microscopy and DIC images of permeabilized osteoclasts treated or not treated (control) with 100 μm (1h) NAC and stained with an α-La antibody that recognizes oxidized La at 3 days post RANKL application. (b) Quantification of a. (n=5) (p=0.009). (c) Fluorescence microscopy and DIC images of permeabilized osteoclasts treated or not treated (control) with 100 μm (1h) NAC and stained with an α-La antibody that recognizes reduced La at 3 days post RANKL application. (d) Quantification of c. (n=2) (p=0.04). Statistical significance evaluated via paired t-test.