Figure 1. Ca ²⁺ -binding Syt1 is required for migrasome formation. (A) Confocal image of migrasomes of L929 cells treated with 0, 2.5, 5, and 10 μM BAPTA and cell migrate on chamber overnight for 12 h. L929 are labeled with wheat germ agglutinin (WGA) (green). Scale bar, 10 μm. (B) The number of migrasomes per 100 μm RF length per cell from images like those shown in A was quantified. Data shown represent the mean ± SEM; P values were calculated using a two-tailed, unpaired t test. n = 50 cells per group. The experiments were performed three times. (C) Confocal image of migrasomes from NRK and NRK Syt1-GFP. Cells are labeled with WGA (magenta). Scale bar, 10 µm or 2 µm (inset). (D) Quantification of the number of migrasomes per 100 μm RF length per cell in NRK and NRK Syt1-GFP cells shown in C. Data represent the mean ± SEM. P values were calculated using a two-tailed, unpaired t test. n = 50 cells per group. The experiments were performed three times. (E) Quantification of the Max Feret’s diameter of migrasomes per cell in NRK and NRK Syt1-GFP cells shown in C. Data represent the mean ± SEM. P values were calculated using a two-tailed, unpaired t test. n = 50 cells per group. The experiments were performed three times. (F) Confocal image of migrasomes from NRK Syt1-GFP treated with/without 1 mM Ca²⁺ for 12 h. Scale bar, 10 µm. (G) Quantification of the number of migrasomes per 100 μm RF length per cell in NRK Syt1-GFP cells shown in F. Data represent the mean ± SEM. P values were calculated using a two-tailed, unpaired t test. n = 50 for NRK Syt1-GFP; n = 50 for NRK Syt1-GFP treated with Ca²⁺. The experiments were performed three times. (H) Confocal image of migrasomes from NRK, NRK Syt1-GFP, and NRK Syt1-GFP treated with 5 μM BAPTA for 12 h. NRK are labeled with WGA (magenta). Scale bar, 10 μm. (I) The number of migrasomes per 100 μm RF length per cell from H was quantified. Data shown represent the mean ± SEM. P values were calculated using a two-tailed, unpaired t test. n = 50 cells per group. The experiments were performed three times. (J) Schematic representation of Syt1 showing transmembrane domain, C2A domain, C2B domain, the phospholipid binding region, the calcium-binding sites, and the sites on the tips of membrane-binding loops. (K) Confocal image of migrasomes from NRK, NRK Syt1-mCherry, NRK Syt1(△C2A)-mCherry, NRK Syt1(△C2B)-mCherry, and NRK Syt1(△C2AB)-mCherry. The △C2A represents the truncation of C2A, the △C2B represents the truncation of C2B. The △C2AB represents the truncation of the phospholipid binding region. Cells are labeled with WGA (green). Scale bar, 10 µm. (L) Quantification of the number of migrasomes per 100 μm RF length per cell in NRK, NRK Syt1-mCherry, NRK Syt1(△C2A)-mCherry, NRK Syt1(△C2B)-mCherry, and NRK Syt1(△C2AB)-mCherry cells shown in K. Data represent the mean ± SEM. P values were calculated using a two-tailed, unpaired t test. n = 51 cells per group. The experiments were performed three times. (M) Confocal image of migrasomes from NRK, NRK Syt1-GFP, NRK Syt1(C2A*)-GFP, NRK Syt1(C2B*)-GFP, and NRK Syt1(C2A*B*)-GFP. The C2A* represents the calcium—binding site mutant (D178A, D230A, and D232A) on C2A, the C2B* represents the calcium—binding site mutant (D309A, D363A, and D365A) on C2B, the C2A*B* represents the calcium—binding site mutant (D178A, D230A, D232A, D309A, D363A, and D365A) on C2AB. Cells are labeled with WGA (magenta). Scale bar, 10 µm. (N) Quantification of the number of migrasomes per 100 μm RF length per cell in NRK, NRK Syt1-GFP, NRK Syt1(C2A*)-GFP, NRK Syt1(C2B*)-GFP, and NRK Syt1(C2A*B*)-GFP cells shown in M. Data represent the mean ± SEM. P values were calculated using a two-tailed, unpaired t test. n = 50 cells per group. The experiments were performed three times. (O) Confocal image of migrasomes from NRK Syt1-GFP and NRK Syt1(4A)-GFP. The Syt1(4A) represent the mutant on the tips of the Syt1 membrane-binding loops (M173A, F234A, V304A, and I367A). Scale bar, 10 µm. (P) Quantification of the number of migrasomes per 100 μm RF length per cell in NRK Syt1-GFP and NRK Syt1 (4A)-GFP shown in O. Data represent the mean ± SEM. P values were calculated using a two-tailed, unpaired t test. n = 51 cells per group. The experiments were performed three times. More about this image found in Ca ²⁺ -binding Syt1 is required for migrasome formation. (A) ...

Figure 2. Syt1 promotes the formation of migrasome-like structures in vitro. (A) Recombinant Syt1-GFP was analyzed on an SDS-PAGE gel and stained with Coomassie Brilliant Blue. (B) Control GUVs and GUVs embedded with Syt1-GFP treated with or without Ca²⁺ were subjected to the in vitro reconstitution assay. Green signal, Syt1-GFP. Magenta signal, Rhodamine-PE. Scale bar, 2 µm. (C) Quantification of the number of migrasome-like structures per 100 µm of the tether from images in B. Data shown represent the mean ± SEM; P values were calculated using a two-tailed, unpaired t test. n = 102 (no Syt1, no Ca²⁺), 108 (no Syt1, with Ca²⁺), 172 (with Syt1, no Ca²⁺), and 213 (with Syt1, with Ca²⁺) tethers. Quantifications are pooled from three independent experiments. Each biological replicate is color-coded: the statistical data from one experimental run is red, another independent experiment is represented by gray, and a third experiment is shown as blue. The dots, squares, triangles, and inverted triangles represent the four groups respectively. The bordered shapes represent those three means. (D) Image of Syt1-containing GUVs in vitro reconstitution assay. The mean intensity was generated using ImageJ. Scale bar, 2 µm. (E) Mean fluorescence intensity of Syt1 along the white lines in D. Source data are available for this figure: SourceData F2 . More about this image found in Syt1 promotes the formation of migrasome-like structures in vitro...

Figure 3. The recruitment of Syt1 induces an unstable swelling of migrasomes prior to the recruitment of TSPAN4. (A) Time-lapse imaging of NRK cells stably expressing Syt1-mCherry and TSPAN4-GFP. Confocal microscopy images were captured every 2 min 54 s. Scale bar, 2 μm. (B) Statistical analysis of normalized fluorescence intensity of Syt1 and TSPAN4 at migrasome formation sites during migrasome formation shown in A. Data are presented as mean ± SEM; n = 30. (C) Time-lapse imaging of NRK Syt1-GFP migrasome formation. Green signal, Syt1-GFP. The time-lapse imaging from top to bottom represent three independent migrasomes. Scale bar, 2 µm. (D) Time-lapse imaging of migrasomes from NRK Syt1-mCherry, NRK TSPAN4-GFP, and NRK Syt1-mCherry TSPAN4-GFP. Scale bar, 2 µm. (E) Statistical analysis of the shrinking back migrasomes to all migrasomes ratio shown in D. Data are presented as mean ± SEM; P values were calculated using a two-tailed, unpaired t test. n = 50. The experiments were performed three times. (F) Quantification of the life time of migrasome shown in D. Data represent the mean ± SEM. P values were calculated using a two-tailed, unpaired t test. n = 50 cells per group. The experiments were performed three times. (G) Time-lapse imaging of migrasomes from NRK Syt1-GFP treated with/without 1 mM Ca²⁺ for 12 h. Scale bar, 2 µm. (H) Quantification of the life time of migrasome shown in G. Data represent the mean ± SEM. P values were calculated using a two-tailed, unpaired t test. n = 100 cells per group. The experiments were performed three times. More about this image found in The recruitment of Syt1 induces an unstable swelling of migrasomes prior to...

Figure 4. Syt1 is necessary for migrasome biogenesis. (A) Confocal images of migrasomes of cells derived from WT and Syt1 knockout L929. Cells are labeled with WGA (cyan). Scale bar, 10 µm. (B) Quantification of the number of migrasomes in WT and Syt1 knockout L929 shown in A. Data represent the mean ± SEM. P values were calculated using a two-tailed, unpaired t test. n = 50 cells per group. The experiments were performed three times. (C) Western blot analysis of Syt1 and GAPDH in WT and Syt1 knockout L929. (D) Confocal images of migrasomes of cells derived from WT and Syt1 knockdown MiaPaCa2. Cells are labeled with WGA (cyan). Scale bar, 10 µm. (E) Quantification of the number of migrasomes in WT and Syt1 knockdown MiaPaCa2 shown in D. Data represent the mean ± SEM. P values were calculated using a two-tailed, unpaired t test. n = 50 cells per group. The experiments were performed three times. (F) Western blot analysis of Syt1 and GAPDH in WT and Syt1 knockdown MiaPaCa2. (G) Confocal images of migrasomes of cells from WT and Syt1 knockout L929 with 0.5 μM BAPTA pretreatment and following with/without 1 mM Ca²⁺ rescue. Left panels: Cells cultured overnight without BAPTA pretreatment. Middle panels: Cells underwent BAPTA pretreatment overnight without Ca²⁺ rescue after 6 h. Right panels: Cells experienced BAPTA pretreatment, 6 h after BAPTA pretreatment, Ca²⁺ was added into culture medium to substitute for the medium. Cells were labeled with WGA (green). Scale bar represents 10 µm. (H) Quantification of the number of migrasomes per 100 μm RF length per cell in WT and Syt1 knockout L929 shown in G. Data represent the mean ± SEM. P values were calculated using a two-tailed, unpaired t test. n = 50 cells per group. The experiments were performed three times. (I) Confocal image of migrasomes of mouse embryonic stem cells treated with N2B27 at D0 and D4. Cells are labeled with WGA (green). Scale bar, 10 µm. (J) The number of migrasomes per cell from images like those shown in I was quantified. Data shown represent the mean ± SEM; P values were calculated using a two-tailed, unpaired t test. n = 50 cells per group. The experiments were performed three times. (K) mRNA levels of Syt1, determined by quantitative PCR analysis, of D0-D5 in mouse embryonic stem cells treated with N2B27. Quantitative PCR data were normalized to Gapdh mRNA level and data are reported as the mean ± SEM. n = 3 independent experiments. (L) Confocal images of migrasomes of cells derived from WT and Syt1 knockout mouse embryonic stem cells treated with N2B27 at D4. Cells are labeled with WGA (cyan). Scale bar, 10 µm. (M) Quantification of the number of migrasomes in WT and Syt1 knockout group shown in L. Data represent the mean ± SEM. P values were calculated using a two-tailed, unpaired t test. n = 50 cells per group. The experiments were performed three times. (N) Western blot analysis of Syt1 and GAPDH in WT and Syt1 knockout mouse embryonic stem cells treated with N2B27 at D4. (O) Model of the role of Syt1 in migrasome biogenesis. Source data are available for this figure: SourceData F4 . More about this image found in Syt1 is necessary for migrasome biogenesis. (A) Confocal images of migraso...

Figure 1. Ectopic expression of forward and reverse minor satellite transcripts leads to chromosome segregation defects in ESCs. (A and B) RT-qPCR analysis of total (A) and forward and reverse (B) MinSat transcripts in G1-, S-, and G2/M-phase populations of ESCs harboring the Fucci cell cycle reporter system. The schematics on top indicate the position of the PCR primers used and the strategy for the analysis of strand-specific transcripts. Bars, mean value of N = 3 independent experiments (individual dots), normalized to β-actin mRNA and control sample; error bars, standard deviation; P values * ≤ 0.05, ** ≤ 0.01. We note that reverse transcripts consistently displayed lower Ct values (∼24.5) than forward transcripts (∼26). (C) α-Tubulin immunofluorescence and DAPI staining of ESCs expressing one repeat MinSat RNAs. Ctrl, empty pCAG vector control. Scale bars, 5 µm. (D) Percentage of chromosome missegregation events after MinSat expression in total anaphase to telophase cells. Bars, mean; error bars, standard deviation; pair-wise comparisons with control (empty pCAG vector) P values * ≤ 0.05, ** ≤ 0.01, *** ≤ 0.001. The number of biological replicates (N) and the total number of mitotic figures analyzed (n) are indicated. (E) Growth curves of ESCs after MinSat RNA expression. Lines indicate the mean of three independent biological replicates; error bars, standard deviation; P value ** ≤ 0.01. (F) Timeline diagram for sample collection and representative immunoblot for the chromatin fraction of ESCs after expression of one or five repeats of forward or reverse MinSat RNAs, or empty pCAG vector control (Ctrl). Cells were synchronized by colcemid treatment before harvesting as indicated. The WB was repeated three or four times with independent lysates, corresponding to the datapoints on the quantification (G). (G) Immunoblot quantification. Values are fold changes normalized to histone H3 and control group in each individual replicate. Bars, mean; error bars, standard deviation; * P value ≤ 0.05. Statistical test: ANOVA. (H) Representative immunoblot for the total lysate of ESCs after expression of one or five repeats of forward or reverse MinSat RNAs, or empty pCAG vector control (Ctrl). Cells were synchronized by colcemid treatment before harvesting as indicated. The WB was repeated three or four times with independent lysates, corresponding to the datapoints on the quantification (I). (I) Immunoblot quantification. Values are fold changes normalized to GAPDH and control group in each individual replicate. Bars, mean; error bars, standard deviation. Statistical test: ANOVA. Source data are available for this figure: SourceData F1 . More about this image found in Ectopic expression of forward and reverse minor satellite transcripts leads...

Figure 2. Depletion of forward or reverse MinSat transcripts impairs chromosome segregation. (A–C) RT-qPCR analysis of total (A), forward (B), and reverse (C) MinSat transcripts in mouse ESCs transfected with ASO against forward (green), reverse (orange), or both strands (purple), respectively. Scramble (gray), non-targeting ASO control; bars, mean values of N = 3 independent experiments (dots), normalized to β-actin mRNA and control samples; error bars, standard deviation; P values * ≤ 0.05, ** ≤ 0.01, *** ≤ 0.001. (D) Representative images showing ESCs stained with α-tubulin and DAPI. Cells were transfected with ASOs against MinSat RNAs. Scale bars, 5 µm. (E) Quantification of chromosome missegregation events in mouse ESCs transfected with the indicated ASOs. Bars, mean percentage of aberrant chromosome segregation (dots indicate individual experiments); error bars, standard deviation; P values * ≤ 0.05 (pair-wise comparisons to control samples). The number of biological replicates (N) and total number of mitotic events analyzed (n) is indicated. (F) Representative images showing ES cells stained with α-tubulin and DAPI. Cells were transfected with ASOs against MinSat RNAs. The presence of micronuclei was analysed in interphase cells. Scale bars, 5 µm. (G) Quantification of micronuclei in mouse ESCs transfected with the indicated ASOs. Bars indicate the mean percentage of micronuclei per cell (dots indicate individual experiments); error bars, standard deviation; P values ** ≤ 0.01, *** ≤ 0.001 (pair-wise comparisons to control samples). The number of biological replicates (N) and the total number of cells analyzed (n) is indicated. (H) Representative immunoblot of the chromatin fraction of ESCs after ASO-mediated knockdown of MinSat RNAs and colcemid synchronization using antibodies to detect the indicated proteins. Scramble, non-targeting ASO control. (I) Immunoblot quantification. Values are fold changes normalized to histone H3 and control in each individual replicate. Bars, mean of N = 5 independent biological replicates (dots); error bars, standard deviation. Statistical test: ANOVA. (J) Representative immunoblot of the total lysate of ESCs after ASO-mediated knockdown of MinSat RNAs and colcemid synchronization using antibodies to detect the indicated proteins. Scramble, non-targeting ASO control. (K) Immunoblot quantification. Values are fold changes normalized to GAPDH and controlled in each individual replicate. Bars, mean of N = 3 independent biological replicates (dots); error bars, standard deviation. Statistical test: ANOVA. (L) Immunofluorescent intensity quantification of CENPC. Data points are means of all sampled CENPC intensity normalized with DAPI signal in that specific Z plane. Scramble, non-targeting ASO control. Bars, mean of N = 3 independent biological replicates (dots); error bars, standard deviation. Source data are available for this figure: SourceData F2 . More about this image found in Depletion of forward or reverse MinSat transcripts impairs chromosome segre...

Figure 3. Minor satellite transcripts interact with RNA-binding regions of CENPC and are structured. (A) Immunoblot for CENPC after RNA pull-down from whole cell lysate of ESCs using biotin-labeled RNAs. The biotinylated RNAs used are indicated on top of the gel image, EGFP was used as negative ... More about this image found in Minor satellite transcripts interact with RNA-binding regions of CENPC and ...

Figure 4. The CENPB box of both mouse MinSat and human ASAT RNAs folds into highly similar stem-loop structures. (A) Sequences of 1 repeat mouse MinSat RNA and human ASAT RNA corresponding to the human chromosome 21 (chr21) α-satellite. The blue underlined region corresponds to the CENPB box. (B) Apical loop structures of MinSat RNA and human chr21 ASAT RNA based on SHAPE analysis. Note that in both species the CENPB box motif (blue) are embedded in the apical loop structures. (C) EMSA of the predicted RNA binding regions in CENPC and 1 repeat of forward human chr21 ASAT RNA. Protein concentrations are indicated on the top. (D) Representative images of mouse ESCs stained with α-tubulin and DAPI under the indicated transfection conditions. Ctrl, empty vector control. Scale bars, 5 µm. (E) Quantification of chromosome missegregation events during mitosis in mouse ESCs transfected with a single repeat of human chr21 ASAT RNA. Ctrl, empty vector control. Bars indicate the mean percentages; error bars, standard deviation; *, P value ≤ 0.05 (pair-wise comparison to control). The number of biological replicates (N) and the total number of mitotic figures analyzed (n) are indicated. Source data are available for this figure: SourceData F4 . More about this image found in The CENPB box of both mouse MinSat and human ASAT RNAs folds into highly si...

Figure 5. Structural conservation rather than sequence motifs determines effects o f MinSat RNAs on chromosome segregation. (A) Sequence comparison of the various MinSat RNA constructs used: wild type: single repeat of consensus forward MinSat RNA sequence; Deletion: deletion of central loop sequence; GNRA tetraloop: replacement of the loop with a smaller stable tetraloop sequence; Swap: swapped CG base pairing in the stem region. The yellow highlighted region indicates the CENPB box motif, blue-colored nucleotides are an integral part of the CENPB box motif and red-colored nucleotides are positions that were mutated within the CENPB box motif in our constructs. (B) Predicted structures of wild-type, deletion, GNRA tetraloop, and swap RNAs using mFold ( Zuker, 2003 ). The yellow highlighted region depicts the CENPB box motif in wild-type sequence and red-colored nucleotides correspond to the mutated positions as depicted in A. (C) Quantification of MinSat RNA levels in ESCs after transfection of the pCAG plasmids containing different mutants of MinSat using RT-qPCR analysis. Bars, mean values of N = 3 independent experiments (dots) normalized to β-actin mRNA and empty pCAG vector control (Ctrl); error bars, standard deviation. (D) Representative images of mouse ESCs transfected with plasmids expressing empty plasmid, wild-type, or different mutants. Cells were immunostained with α-tubulin and DAPI. Scale bars, 5 µm. (E) Quantification of chromosome missegregation events in anaphase and telophase mouse ESCs after overexpression of the structural RNA mutants, wild-type, or empty pCAG vector as control (Ctrl), respectively. Bars indicate the mean percentage of N = 3 independent experiments (dots); error bars, standard deviation; *, P value ≤ 0.05 (pair-wise comparison to control). More about this image found in Structural conservation rather than sequence motifs determines effects o f...

Figure 1. Clogger removal from mitochondrial import channels in mammals and yeast. Molecules involved in clogger removal and representative cloggers utilized in each pathway are shown. The mammalian version of clogger developed by Krakowczyk et al. ( 2 ) is integrated in the IMM component, ATP s... More about this image found in Clogger removal from mitochondrial import channels in mammals and yeast. M...