Autophagy machinery mediates macroendocytic processing and entotic cell death ... - Weebly · 2021. 2. 1. · DIC mCherryÐLC3 GFPÐLC3 G120A Merge 0 150 300 X X X X X Time (min)

ART I C L E S

Autophagy machinery mediates macroendocyticprocessing and entotic cell death by targeting singlemembranesOliver Florey1, Sung Eun Kim1,2, Cynthia P. Sandoval3, Cole M. Haynes1,2 and Michael Overholtzer1,2,4

Autophagy normally involves the formation of double-membrane autophagosomes that mediate bulk cytoplasmic and organelledegradation. Here we report the modification of single-membrane vacuoles in cells by autophagy proteins. LC3 (Light chain 3) acomponent of autophagosomes, is recruited to single-membrane entotic vacuoles, macropinosomes and phagosomes harbouringapoptotic cells, in a manner dependent on the lipidation machinery including ATG5 and ATG7, and the class IIIphosphatidylinositol-3-kinase VPS34. These downstream components of the autophagy machinery, but not the upstreammammalian Tor (mTor)-regulated ULK–ATG13–FIP200 complex, facilitate lysosome fusion to single membranes and thedegradation of internalized cargo. For entosis, a live-cell-engulfment program, the autophagy-protein-dependent fusion oflysosomes to vacuolar membranes leads to the death of internalized cells. As pathogen-containing phagosomes can be targeted ina similar manner, the death of epithelial cells by this mechanism mimics pathogen destruction. These data demonstrate thatproteins of the autophagy pathway can target single-membrane vacuoles in cells in the absence of pathogenic organisms.

The proper degradation of material taken into a cell by macroscaleendocytic processes1 is critical for a range of metazoan cell functionsincluding erythroblast enucleation2, axon pruning3, removal ofdying cells4, antigen presentation5 and the clearance of pathogenicorganisms6. Engulfed cargo is directed towards degradation by acomplex series of lipid phosphorylation and protein recruitment eventsthat direct phagosome fusion with lysosomes7. Critical among theseis the recruitment of the lipid kinase VPS34, and accumulation of itsproduct phosphatidylinositol-3-phosphate (PtdIns(3)P), that initiatesmaturation through stages marked by activation of the small GTPasesRAB5 and RAB7, which direct vesicular fusion8.Macroautophagy (commonly termed autophagy) is another

lysosomal delivery pathway, which mediates the formation ofdouble-membrane autophagosomes that enwrap cellular componentsfor delivery to the lysosome9. A signalling complex involving ULK1,ATG13 and FIP200 is responsible for activating two ubiquitin-likeconjugation systems, controlled by autophagy (ATG) proteins(ATG3, 4, 5, 7, 12 and 16), that form autophagosomes in part byconjugating LC3 to phosphatidylethanolamine10. The recovery ofamino acids and other building blocks by autophagy is widely viewedas a survival mechanism for cells undergoing starvation11–13. Theautophagy pathway also targets a variety of pathogenic organisms for

1Cell Biology Program, Memorial Sloan Kettering Cancer Center, New York, New York 10065, USA. 2BCMB Allied Program, Weill Cornell Medical College, 1300 YorkAvenue, New York, New York 10065, USA. 3Department of Physiology, University of Arizona, Tucson, Arizona 85721, USA.4Correspondence should be addressed to M.O. (e-mail: [email protected])

Received 9 March 2011; accepted 16 September 2011; published online 16 October 2011; DOI: 10.1038/ncb2363

degradation by sequestration into double-membrane autophagosomes,including Listeria monocytogenes14,15, Mycobacterium tuberculosis16,Streptococcus pyogenes17, Shigella18 and Toxoplasma gondii19. However,for Escherichia coli and Saccharomyces cerevisiae, autophagy pathwayproteins have been shown to facilitate killing and degradation bydirectly conjugating LC3 to phagosome membranes, in a mannerindependent of double-membrane autophagosomes, supporting anon-canonical role for autophagy proteins in innate immunity, referredto as LC3-associated phagocytosis20,21.Recently, a cell death mechanism called entosis was reported22.

Entosis, similarly to phagocytosis, is a formof cell engulfment.However,unlike phagocytosis, which targets dead or dying cells, entosis occursbetween live epithelial cells. Internalizing cells play an active rolein engulfment, which results in complete internalization within an‘entotic vacuole’, whose membrane is derived from invagination ofthe host-cell plasma membrane. The cell-in-cell structures resultingfrom entosis are found in human cancers, but their role remainsunknown23. Although some entotic cells can escape from their hosts,most undergo cell death, indicating that entosis could be a mechanismof tumour suppression22. Entotic cell death is non-apoptotic andinvolves lysosomal acidification of the entotic vacuole; however, themolecular mechanism has not been defined.

NATURE CELL BIOLOGY VOLUME 13 | NUMBER 11 | NOVEMBER 2011 1335

© 2011 Macmillan Publishers Limited. All rights reserved.

mailto:[email protected]

ART I C L E S

Here we report the mechanism of cell death by entosis. We findthat entotic vacuole membranes surrounding internalized cells recruitthe autophagy protein LC3, in a manner dependent on autophagymachinery including ATG5, ATG7 and VPS34, but independentof autophagosome formation. LC3-targeted entotic vacuoles recruitlysosomes, resulting in the degradation of internalized cells, which arekilled by their hosts. These data define amammalian cell death programthat shares characteristics with pathogen destruction21, and thatrequires autophagy proteins. Remarkably, we also find that apoptoticcell phagosomes in macrophages, andmacropinosomes, are targeted byLC3, in a manner dependent on autophagy proteins, but independentof autophagosomes. These data demonstrate that the targeting ofsingle-membrane vacuoles is a general function of autophagy pathwayproteins that is not restricted to the innate immune system.

RESULTSLC3 is recruited to single-membrane entotic vacuoles in anautophagosome-independent mannerDuring entosis, viable cells are internalized inside host cells forextended periods—from one to many hours—before most undergoa non-apoptotic form of cell death22. As the autophagy pathway maycontribute to cell death in some contexts, we examined autophagyduring entotic death by time-lapse imaging the autophagosomemarker GFP–LC3. Surprisingly, we discovered a rapid and transientrecruitment of GFP–LC3 from host cells onto entotic vacuolemembranes22 (Fig. 1a and Supplementary Movie S1 and Fig. S1a).Lipidation of LC3 to phosphatidylethanolamine was required forrecruitment, as GFP–LC3G120A, a mutant incapable of lipidation, didnot recruit to the entotic membrane, whereas mCherry–LC3 imagedin the same cell did (Fig. 1b).To examinewhether LC3was recruited by autophagosome fusion, we

imaged live cells by four-dimensional confocal microscopy. Strikingly,increased GFP–LC3 intensity at vacuole membranes was mirroredby a loss of cytoplasmic fluorescence signal in host cells (Fig. 1c andSupplementary Movie S2), but not by a change in the number ofGFP–LC3 puncta (Supplementary Fig. S1b). We did not observe overtmovement of GFP–LC3 puncta to entotic vacuoles, which mightbe expected if LC3 recruited by autophagosome fusion. Next weexamined the GFP–LC3-loaded entotic vacuole by correlative video-light–electron microscopy to interrogate its structure. A cell-in-cellstructure was imaged by time-lapse microscopy and fixed after theearliest observable recruitment of GFP–LC3 (Fig. 1d). Examinationof the same cells by transmission electron microscopy revealed asingle-membrane structure of the GFP–LC3-loaded entotic vacuole.There were no observable autophagosome contents within the vacuolelumen, which would be expected on fusion of double-membraneautophagosomes (Fig. 1d and Supplementary Fig. S1c and Movie S3).These data are consistent with a model whereby GFP–LC3 is directlyrecruited from the cytosolic pool to the single-membrane entoticvacuole, where it is lipidated.To explore the requirement of autophagy machinery for LC3

recruitment, short interfering RNAs (siRNAs) targeting expression ofautophagy proteins, as well as the VPS34 inhibitor 3-methyladenine24

(3-MA), were introduced into cells. VPS34, ATG5 and ATG7 arerequired for LC3 lipidation25–27, whereas FIP200 is a component of theupstream ULK complex that initiates autophagosome formation28–30.

The knockdown of VPS34, ATG5 or ATG7 significantly reduced thepercentage of entotic vacuoles that recruited LC3, demonstratinga requirement for lipidation machinery and VPS34 (Fig. 1e andSupplementary Fig. S2a). In contrast, knockdown of FIP200, whichblocked starvation-induced autophagy similarly to knockdown ofATG5 (Supplementary Fig. S2d,f,g), had no effect on LC3 recruitment(Fig. 1e). These data distinguish single-membrane targeting frommacroautophagy, and support a model whereby LC3 recruitment toentoticmembranes occurs independently of autophagosomes.

Lysosome fusion to entotic vacuoles and internalized cell deathTo further characterize thematuration of entotic vacuoles, we examinedevents upstream and downstream of LC3. The VPS34 product PI(3)Precruits FYVE-domain proteins that function upstream of LC3 inautophagy27,31,32. A fluorescent reporter of PI(3)P, 2×FYVE–mCherry,recruited to entotic vacuoles approximately 12min (11.9min ±4.3) before LC3 (Fig. 1f and Supplementary Movie S4), indicatingthat PI(3)P production at the vacuole membrane could directLC3 lipidation. We also imaged cells expressing the lysosomemarker Lamp1–GFP in combination with mCherry–LC3. Lamp1–GFPrecruited to entotic vacuoles approximately 30min after LC3, but stillbefore the death of internalized cells (Fig. 1g and SupplementaryMovieS5). By expressing mCherry-tagged Cathepsin B we also confirmed thatlysosomal enzymes from the host cell deposit into the entotic vacuolebefore internalized cell death (Fig. 1h). The recruitment of LC3 andconversion of entotic vacuoles to lysosome compartments before celldeath indicates that entotic cellsmay be killed by their hosts (Fig. 1i,j).

Autophagy lipidation machinery and VPS34 are required for LC3recruitment and non-apoptotic ‘killing’ of internalized cells.Wenext examinedwhether LC3 recruitment is required for internalizedcell death. By quantifying cell fate following knockdown of VPS34,ATG5 or ATG7, we found a small but significant decrease in the levelof internalized cell death (Fig. 2b and Supplementary Fig. S2b). Incontrast, depletion of FIP200, which had no effect on LC3 recruitment(Fig. 1e), also had no effect on cell fate (Fig. 2b). To localize the re-quirement of autophagy machinery, cell fates were quantified in mixedcell-in-cell structures, withATG5 inhibited in either internalized cells orhosts. Knockdown of ATG5 in hosts but not internalized cells inhibitedLC3 recruitment and partially rescued internalized cells from death(Fig. 2c,d), demonstrating that autophagy lipidation machinery in hostcells is required for LC3 recruitment and internalized cell killing.

Internalized cells require macroautophagy to survive withinentotic vacuolesOverall, inhibition of internalized cell death by knockdown ofautophagy machinery (Fig. 2b) was modest, considering the morethan threefold inhibition of LC3 recruitment by depleting ATG5or ATG7 (Fig. 1e). By monitoring cell death morphology with thenuclear marker H2B–mCherry, it was evident that autophagy-inhibitedinternalized cells were undergoing apoptosis, whereas control cells diednon-apoptotically (Fig. 2e). Importantly, knockdown of either FIP200or ATG5 significantly increased the frequency of apoptosis, linkingthis phenotype to macroautophagy inhibition. This is reminiscent ofprevious results where lysosome inhibition switched entotic cell deathfrom non-apoptotic to apoptotic22. We reasoned that internalized cells

1336 NATURE CELL BIOLOGY VOLUME 13 | NUMBER 11 | NOVEMBER 2011


ART I C L E S

Inner cell

Host

0:18:01 0:36:02 0:48:01 2:27:50DIC

H2

B–m

Ch

err

y

GF

P–L

C3

x–z

4 μm

EM of same cell

Phase GFP–LC3

150 nm

Time (min)

Flu

ore

scen

ce in

ten

sity

fo

ld c

han

ge

∗∗∗

∗∗∗∗∗∗

Control VPS34siRNA

Control ATG5siRNA

FIP200siRNA

Control ControlATG7siRNA

DIC mCherry–LC3

GFP–LC3G120A Merge

0 150 300

X

XX

X

X

Time (min)

450

Time (min )

2×F

YV

E–m

Ch

err

y

GF

P–L

C3

Lam

p1

–G

FP

Cath

ep

sin

-B–m

Ch

err

y

DIC

*

*

Vacuole 1Vacuole 2Cytoplasm 1Cytoplasm 2

0:45:44 0:51:43 1:03:44 1:27:44DIC 21 min 39 min 51 min 102 minDIC

0

0.5

1.0

1.5

2.0

2.5

0 20 40 60 80

P

erc

en

tag

e o

f cell

death

s w

ith

LC

3 r

ecru

itm

en

t

PI(3)P

LC3

Lamp1

0 50 100 150

a

f

h i j

g

d e

b c

mC

herr

y–L

C3

0

25

50

75

100

n = 40 n = 21 n = 61 n = 49 n = 49 n = 45n = 108 n = 103

Figure 1 Entotic cell death involves recruitment of LC3 to a single-membranevacuole. (a) Confocal time-lapse images of an MCF10A cell-in-cell structureexpressing GFP–LC3 and H2B–mCherry. The arrow marks recruitment ofLC3 to the entotic vacuole (see Supplementary Movie S1). Scale bar, 15 µm.(b) Images of MCF10A cells expressing GFP–LC3G120A and mCherry–LC3;note no recruitment of GFP–LC3G120A to the entotic vacuole. Scale bar,15 µm. (c) LC3 is recruited from the host-cell cytosol onto the entoticvacuole membrane. Data points are the mean fluorescence intensity ofGFP–LC3 on vacuoles versus cytoplasm for two independent cell-in-cellstructures (see Supplementary Movie S2). (d) Correlative video-light–electronmicroscopy of an MCF10A GFP–LC3 cell-in-cell structure. Cells were imagedby time-lapse microscopy, followed by fixation after LC3 recruitment (seeSupplementary Movie S3). Top left, post-fixation images of phase-contrastand GFP-fluorescence signals were taken demonstrating LC3 recruitment.Bottom left and right, electron microscopy (EM) of the same cell shows theentotic vacuole is a single membrane (arrow in right image, which is a highermagnification of the area outlined in the bottom left image). (e) Percentageof MCF10A cell-in-cell structures, with and without autophagy inhibition,

that show LC3 recruitment associated with death of the internalized cell.Data represent mean±s.e.m. from at least three independent experiments;n, total number of death events; ∗∗∗P < 0.0001. (f,g) Time-lapse imagesof MCF10A cell-in-cell structures expressing either 2×FYVE–mCherry (red)and GFP–LC3 (green; f) or Lamp1–GFP (green) and mCherry–LC3 (red; g).The arrows indicate recruitment to the entotic vacuole (see SupplementaryMovies S4 and S5). Times are indicated as h:min:s (a,f) or min (g). Scalebars, 5 µm. (h) Representative images of cell-in-cell structures in whichonly host cells express Cathepsin-B–mCherry. Left, a live inner cell withno Cathepsin B in the vacuole (arrow); middle, a live internalized cell withCathepsin B from the host inside of the vacuole (arrowhead); right, a deadinternalized cell with Cathepsin B from the host throughout the corpse(asterisk). Scale bar, 15 µm. (i) Five representative timings of GFP–LC3recruitment to entotic vacuoles (green bar) and death of internalized cells(cross) from time-lapse microscopy. (j) Initiation and duration of 2×FYVEdomain (red bar, n =11), LC3 (green bar, n =20) and Lamp1 (orange bar,n=7) recruitment to entotic vacuoles, determined using double-expressingcells shown in f,g. Error bars represent s.d.



ART I C L E S

DeathReleaseNo change

∗∗

∗

Perc

en

tag

e o

f cell

fate

Perc

en

tag

e o

f cell

fate

∗

Host-cell siRNA Inner-cell siRNA

∗∗ ∗

Perc

en

tag

e o

f cell

death

s

with

LC

3 r

ecru

itm

en

t

Death

Rele

ase

12

12

1

2

12

1

2 12

H2B–mCherry

Time

∗∗ ∗

∗∗

Control

n = 77 n = 67

∗∗

n = 50

∗∗

DIC

H2

B–m

Ch

err

y

GF

P–L

C3 mC

herr

y/G

FP

Inner cells

∗∗∗∗∗

∗ ∗∗

Control Inner cell

Perc

en

tag

e o

f

au

top

hag

oso

me a

rea

∗∗

300 nm500 nm

Perc

en

tag

e o

f d

eath

DeathRelease

Division/no change

Cell-in-cell

n = 28 n = 25

ATG5siRNA

ATG5siRNA

ATG5siRNA

ATG5siRNAn = 24

Control

n = 63 n = 75 n = 55

Control

n = 53

ControlATG5siRNA

ATG5siRNA

ATG7siRNA

Control Control FIP200siRNA

FIP200siRNA

Control ControlVPS34siRNA

n = 24

Control0

25

50

75

100

0

25

50

75

100

DeathReleaseNo change

ApoptoticNon-apoptotic

0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

Gro

wth

HB

SS

Co

ntr

ol

VP

S34

siR

NA

Co

ntr

ol

ATG

5 siR

NA

Co

ntr

ol

FIP

200

siR

NA

+ V

PS

34 s

iRN

AH

BS

S

+V

PS

34 s

iRN

AG

row

th

0

4

8

12

a

b

e

g h

f

d

c

4 μm

Host-cell siRNA Inner-cell siRNA

0

25

50

75

100

0

25

50

75

100

n = 86 n = 261 n = 149 n = 98 n = 88n = 230 n = 142n = 86

Figure 2 Autophagy machinery in host cells controls the fate of starvinginternalized cells. (a) Left, model depicting the multiple cell fates ofinternalized cells; the arrow thickness represents the relative frequency ofevents. Right, images of MCF10A cell-in-cell structures (1, host cell; 2,internalized cell) that show a cell death (top images) or release (bottomimages) event. Scale bar, 15 µm. (b) Quantification of MCF10A internalizedcell fate over 20h following autophagy inhibition. Data represent the meanfrom at least three independent experiments; n, total number of structuresanalysed; ∗∗∗P < 0.007, ∗∗P < 0.004, ∗P < 0.01. (c,d) Quantification ofLC3 recruitment (c), and cell fate (d), of mixed cell-in-cell structuresfor which host or internalized cells were treated with ATG5 siRNA. Datarepresent mean±s.e.m.from three separate experiments; n, total numberof structures analysed; ∗P <0.01, ∗∗P <0.004. (e) Left, quantification oftype of internalized cell death after autophagy inhibition. Data represent

the mean of three independent experiments; n, total death eventsanalysed; ∗∗P < 0.03. Right, representative images of non-apoptotic andapoptotic death of an internalized cell; the arrow points to fragmentedapoptotic nuclei. Scale bars, 15 µm. (f) Quantification of mCherry/GFPratio of live MCF10A single cells or internalized cells expressing tandemGFP–mCherry–LC3, with or without siRNA treatment as indicated. Datarepresent mean± s.d.; numbers of cells analysed for each data point(from left to right) 30, 30, 30, 25, 30, 25, 25, 25, 30, 27; ∗P < 0.03,∗∗P < 0.003, ∗∗∗P < 0.0004. (g) Quantification of autophagosomeand autolysosome area in control MCF10A (n = 10) and internalizedcell (n = 10) cytoplasm imaged by electron microscopy; ∗∗P < 0.005.(h) Representative electron micrographs of cell-in-cell structures (left) andinternalized cell cytoplasm (middle and right) showing autophagosomes(arrow) and autolysosomes (arrowhead).

undergo starvation inside host-cell vacuoles, and rely on autophagyfor survival. Internalized cells were restricted for access to growthmedia, as assessed by uptake of fluorescently labelled epidermal growth

factor (EGF–Alexa488; Supplementary Fig. S3a), and they had lowlevels of phosphorylated S6 ribosomal protein, similar to starved cells(Supplementary Fig. S3b).



ART I C L E S

Con

trol

3-M

A

ATG5

siRNA

Bcl2

ATG

5

siRNA

Bcl

2 3-

MA

Control ATG5.3 ATG5.9

Control 3-MA Y27632Perc

en

tag

e o

f cell-

in-c

ell

(so

ft

ag

ar

48

h)

Co

lon

y n

um

ber

Rele

ase

(perc

en

tag

e o

f co

ntr

ol) ∗

∗ ∗

H2B–mCherryGFP–LC3DIC

0

500

1,000

1,500

2,000

2,500

Control

siRNA siRNA siRNA siRNA

3-MA Y27632

∗∗

∗ ∗∗

∗∗ Co

lon

y n

um

ber

Control ATG5.3 ATG5.9

∗ ∗

Con

trol

Y2763

2

3-M

A

ATG5

siRNA

VPS34 siR

NAP

erc

en

tag

e o

f cell-

in-c

ell

form

atio

n (su

sp

en

sio

n 8

h)

∗∗

0

5

10

15

20

0

10

20

30

40

50

0

250

500

750

1,000

a

d

f

e

g

b c

0

100

200

300

400

500

Figure 3 Inhibition of apoptosis and entotic cell death promotesanchorage-independent growth. (a) The fate of internalized cells(control MCF10A and MCF10A expressing E7 and Bcl2) was measuredwith and without autophagy inhibition. Data show the percentage ofinternalized cell release with autophagy inhibition, compared witheach control without autophagy inhibition; mean± s.e.m. from atleast three independent experiments (total cell numbers analysed:ATG5 siRNA, 261 cells; Bcl2 ATG5 siRNA, 97 cells; 3-MA, 170cells; Bcl2 3-MA, 112 cells); ∗P < 0.05, ∗∗P < 0.02. (b) Effects ofY27632, 3-MA and VPS34 and ATG5 siRNA on MCF10A cell-in-cellformation in suspension for 7 h; >300 cells per condition were scoredfor cell-in-cell formation; ±s.e.m.from three independent experiments;

∗∗P < 0.004. (c) Representative images of an MCF10A E7+Bcl2cell-in-cell structure formed in soft agar. The arrow marks LC3 recruitmentaround internalized cells. Scale bar, 5 µm. (d) Quantification of MCF10AE7+Bcl2 cell-in-cell structures 48h after seeding into soft agar(white bars), and colony formation after 2 weeks (grey bars). Datarepresent mean±s.e.m.from three independent experiments; ∗P <0.03,∗∗P < 0.004, ∗∗∗P < 0.001. (e) Representative wells for data in d.The insets show higher magnifications. (f) Quantification of MCF10AE7+Bcl2 colony formation with Atg5 knockdown with two separatesiRNAs. Data represent mean±s.e.m.from three independent experiments;∗P <0.001. (g) Representative wells for data in f. The insets show highermagnifications. See also Supplementary Movie S6.

To determine whether internalized cells upregulate autophagy, wequantified red/green fluorescence intensity ratios of cells expressing atandem-tagged LC3 (GFP–mCherry), which allows for quantificationof autophagy on the basis of loss of GFP but retention of mCherryfluorescence signal in lysosomes33. Control cells starved in HBSSexhibited increased red/green ratios, consistent with the activationof autophagy, which was inhibited by VPS34 siRNA. Similarly, inter-nalized cells exhibited increased red/green ratios, which were inhibitedby VPS34, ATG5 or FIP200 knockdown (Fig. 2f). Internalized cellsalso had increased numbers of autophagosomes and autolysosomesidentified by transmission electron microscopy (Fig. 2g,h). Togetherthese data demonstrate that internalized cells are denied access togrowth media and upregulate macroautophagy as a survival response.On macroautophagy inhibition, by siRNA-mediated knockdown ofFIP200 or ATG5, internalized cells are sensitized to apoptosis.

Inhibition of autophagy machinery and apoptosis rescues entoticcells and promotes anchorage-independent growth in soft agarIt was previously reported that Bcl2 overexpression inhibitedconcanamycin-A-induced apoptosis of entotic cells, and thesimultaneous inhibition of apoptosis and lysosome function enabled

internalized cell release22. To achieve a more complete rescue ofinternalized cells in the context of autophagy protein inhibition, weapplied a similar strategy and overexpressed Bcl2 to block apoptosis.The combined inhibition of autophagy proteins and apoptosisdecreased the level of internalized cell death more than autophagyprotein inhibition alone, and significantly increased the percentage ofinternalized cells that escaped from their hosts (Fig. 3a).To examine the potential significance of entotic cell death, we

investigated the effects of autophagy protein inhibition on transformedgrowth in soft agar, where entosis occurs at a high frequency22.We used non-transformed MCF10A cells, which are competent forapoptosis and entosis, and do not grow under anchorage-independentconditions. The pro-proliferative oncogene product HPV-E7, and alsoBcl2, were overexpressed in MCF10A cells to activate proliferationand block apoptosis22. Despite overexpression of these proteins,anchorage-independent growth in soft agar was limited, and entosisoccurred at a high frequency (∼40%, after 48 h). Entotic structuresreadily exhibited cell death (Fig. 3c and Supplementary Movie S6),consistent with the ability of entosis to eliminate cells that are resistantto apoptosis22. We also found that cells could escape from suchstructures, even in soft agar (SupplementaryMovie S6). To examine the



ART I C L E SH

2B

–m

Ch

err

yG

FP

–L

C3

Control

Co

ntr

ol

Atg5 shRNAn = 35 n = 30

x–z

∗∗∗

∗∗∗

∗∗∗

AP

EM of same cell

4 μm 500 nmDIC 0:57:48 3:37:492:13:48 2:53:49

Control FIP200siRNA

Atg5siRNA

n = 6 n = 9n = 14

Perc

en

tag

e o

f cells

with

an

LC

3 r

ing

Perc

en

tag

e o

f p

hag

oso

mes

with

LC

3 r

ecru

itm

en

t

H2

B–m

Ch

err

yG

FP

–L

C3

H2

B–m

Ch

err

yG

FP

–L

C3

Atg5

sh

RN

A

Time (min)

90.5% (n = 21)

23.8% (n = 21)

0:10:00 0:20:00 0:40:00 1:20:00

0

25

50

75

100

0

25

50

75

100

0 20 40 60 80

a e

fc d

b

DIC

Figure 4 LC3 recruits to single-membrane vacuoles containing apoptoticcells to facilitate corpse degradation. (a) Time-lapse images of MCF10Acell-in-cell structures with an internalized cell that undergoes apoptosis.The arrows mark the recruitment of LC3 around apoptotic fragments (seeSupplementary Movie S7). Scale bar, 10 µm. (b) Quantification of LC3recruitment to apoptotic internalized cells in the presence of ATG5 orFIP200 siRNA; n, number of cells analysed; ∗∗∗P <0.0001, chi-squared test.(c) Apoptotic U937 cells expressing H2B–mCherry phagocytized by J774macrophages expressing GFP–LC3. Confocal images show LC3 recruitmentto an engulfed corpse (arrow; see Supplementary Movie S8). Scale bar,

15 µm. (d) Quantification of LC3 recruitment to apoptotic phagosomesin control and Atg5 -shRNA-expressing cells; ∗∗∗P <0.0001, chi-squaredtest. (e) Correlative light–electron microscopy of an apoptotic phagosomewith GFP–LC3 recruitment (top left image, white arrow). The LC3-loadedphagosome consists of a single membrane (black arrow in right image, whichis a higher magnification of the area outlined in the bottom left image; EM,electron microscopy; AP, apoptotic cell nucleus). (f) Representative imagesof apoptotic cell phagocytosis and degradation in control (n=21) and Atg5shRNA (n = 21) J774 GFP–LC3 cells; ∗∗∗P < 0.0001, chi-squared test.Scale bar, 10 µm. See also Supplementary Fig. S3.

role of entosis in this system, we blocked entotic cell death by inhibitingautophagy proteins, or blocked the formation of cell-in-cell structuresby inhibiting Rho-kinase with Y27632, as reported previously22. Theinhibition of entotic death with 3-MA significantly increased thelevel of colony formation, and decreased the frequency of cell-in-cellstructures in soft agar after 48 h (Fig. 3d,e). Unlike Y27632, whichblocks cell-in-cell formation, inhibiting autophagy did not affectthe ability of MCF10A cells to form cell-in-cell structures (Fig. 3b),indicating that the decreased frequency of cell-in-cell in soft agar wasprobably due to its effect on decreasing the level of cell death, whichleads to internalized cell release. In a manner similar to 3-MA, siRNAstargeting ATG5 also increased the level of colony formation (Fig. 3f,g),demonstrating that autophagy proteins act as negative regulators oftransformed growth under these conditions where entosis occurs at ahigh frequency. These data are consistent with entosis, and entotic celldeath, acting as a suppressor of transformed growth.

LC3 is recruited to apoptotic cell phagosomesTo explore the significance of single-membrane modification byautophagy proteins, we considered whether other, non-pathogenengulfments would also be targeted. We noted that entotic vacuoleswere targeted by LC3 even if internalized cells had undergoneapoptosis (Fig. 4a,b and Supplementary Movie S7). To examinewhether macrophages ingesting apoptotic cells would recruit LC3to phagosomes, J774 mouse macrophages expressing GFP–LC3 wereincubated with apoptotic cells expressing H2B–mCherry. Strikingly,GFP–LC3 was rapidly recruited to apoptotic cell phagosomes (Fig. 4cand Supplementary Movie S8). Following LC3 recruitment, corpsesseemed to be degraded rapidly, as evidenced by the release of mCherry

from condensed nuclei, which was blocked by treatment with thelysosome inhibitor concanamycin A (Fig. 4c and SupplementaryFig. S4c and Movie S8). Correlative light–electron microscopy of theLC3-targeted phagosome revealed a single-membrane structure, devoidof fusing vesicles and autophagic bodies, as was observed for theentotic vacuole (Fig. 4e), demonstrating that macrophages recruit LC3to single-membrane apoptotic cell phagosomes. J774 macrophagesexpressing a short hairpin RNA (shRNA) targeting Atg5, whichinhibited autophagy (Supplementary Fig. S4a,b), were significantlyinhibited for recruitment of GFP–LC3 to phagosomes (Fig. 4d).Notably, corpses engulfed by Atg5-shRNA-treated macrophagesresembled those taken in by control macrophages treated withconcanamycin A, as the diffusion of H2B–mCherry, which markslysosome-mediated degradation, was markedly inhibited (Fig. 4f andSupplementary Fig. S5). These data demonstrate a role for autophagyprotein targeting of single-membrane phagosomes in regulating thedegradation of apoptotic corpses.

Macropinosomes recruit LC3In J774 macrophages we also noted the recruitment of GFP–LC3 tomacropinocytic vacuoles. Fifty per cent of macropinosomes rapidlyrecruited GFP–LC3 (Fig. 5a and Supplementary Movie S9) in anAtg5-dependent manner (Fig. 5b). To examine another cell type, weimaged MCF10A, where approximately 50% of macropinosomes alsorecruited GFP–LC3 (Fig. 5c). Recruitment of GFP–LC3 was inhibitedby knockdown of ATG5, but not by the knockdown of FIP200 (Fig. 5d),demonstrating a requirement for autophagy lipidation machinery butindependence from macroautophagy. Interestingly, in line with aprevious report21, we found no recruitment of LC3 to phagosomes



ART I C L E S

Control Atg5 shRNA

n = 21 n = 13

Perc

en

tag

e o

f

macro

pin

oso

mes w

ith

LC

3 r

ecru

itm

en

t

Perc

en

tag

e o

f

macro

pin

oso

mes w

ith

LC

3 r

ecru

itm

en

t

∗∗∗

∗∗∗

LysoTrackerGFP-LC3

DIC 17:23 17:58 18:3418:34

DIC Dextran/GFP–LC3

DIC Dextran /GFP–LC3

0

25

50

75

Control FIP200siRNA

ATG5siRNA

n = 32 n = 40n = 40

ab

dc

e

0

25

50

75

Figure 5 LC3 is recruited to macropinosomes independently ofmacroautophagy. (a) Representative images of a J774 macrophageexpressing GFP–LC3 incubated in media containing red dextran. The arrowmarks LC3 recruitment to a dextran-containing macropinosome. Scalebar, 2 µm. (b) Quantification of LC3 recruitment to macropinosomes incontrol and Atg5 shRNA J774 macrophages; n, number of macropinosomesanalysed; ∗∗∗P <0.0001, chi-squared test. See Supplementary Movie S9.(c) Representative image of LC3 recruitment to macropinosomes (arrows)in MCF10A cells expressing GFP–LC3 incubated with red dextran. Scalebar, 2 µm. (d) Quantification of LC3 recruitment to macropinosomes inMCF10A cells treated with siRNA against FIP200 or ATG5 ; n, numberof macropinosomes analysed; ∗∗∗P < 0.0001, chi-squared test. (e) J774GFP–LC3 macrophages were incubated with 3 µm uncoated latex beadsand imaged by confocal microscopy for 1 h, followed by the addition ofLysoTracker red to mark acidified compartments. A macropinosome recruitedLC3 (arrow) whereas engulfed latex beads did not (arrowheads). Scale bar,5 µm. See Supplementary Movie S10.

surrounding uncoated latex beads, whereas macropinosomes in thesame cell did recruit LC3 (Fig. 5e and Supplementary Movie S10).Latex-bead-containing phagosomes stained positively with Lyso-Tracker, demonstrating their complete engulfment (Fig. 5e).

LGG-1 recruits to apoptotic cell phagosomes duringCaenorhabditis elegans embryo development.To extend our findings to an in vivomodel, we examined whether LC3would recruit to apoptotic cell phagosomes during C. elegans embryodevelopment34,35. By time-lapse imaging embryos co-expressingmCherry::RAB-5, which is known to recruit to phagosomes8, andGFP::LGG-1, the LC3 homologue, we indeed observed recruitment ofLGG-1 to apoptotic cell phagosomes (Fig. 6a,b and SupplementaryFig. S4d and Movie S11). LGG-1 recruited coincidently with, orimmediately after, RAB-5 (Fig. 6b). Depletion of the autophagy proteinBeclin1 homologue, BEC-1, a component of the VPS34 complex,by RNAi significantly decreased the frequency of both LGG-1 andRAB-5 recruitment (Fig. 6c,d). The knockdownof BEC-1 also increased

the number of apoptotic corpses that were observed (Fig. 6e,f), inagreement with a recent report36. This did not seem to be due to a defectin apoptotic corpse uptake, but rather a defect in corpse degradation,as corpses from bec-1 RNAi worms still showed CED-1 recruitment,demonstrating engulfment37 (Fig. 6g). These data are consistent with amodel whereby BEC-1 regulates corpse degradation in C. elegans, atleast in part by controlling phagosome maturation by the recruitmentof RAB-5 and LGG-1.

DISCUSSIONThe data presented here demonstrate targeting of single-membranevacuoles, including entotic vacuoles, apoptotic cell phagosomes andmacropinosomes, by autophagy proteins. Unlike macroautophagy,which is induced by starvation and inactivation of mTOR (ref. 38), thetargeting of single membranes by autophagy proteins occurs in cellsunder normal growth conditions when mTOR is active39. Accordingly,single-membrane targeting of LC3 requires lipidation machinery (forexample ATG5 and ATG7; refs 25,26,40,41), and VPS34, but does notrequire FIP200, a member of the mTOR-inhibited ULK complex thatis required for macroautophagy28–30,42–45.Efficient degradation of phagocytized E. coli or yeast21, the in vivo

resistance to T. gondii46 and presentation of exogenous antigen47

were all recently shown to require autophagy machinery in anautophagosome-independent manner, while relying on the presenceof pathogens or Toll-like receptor ligands. We now show a similar rolefor autophagy proteins in the clearance of apoptotic cells, a processof fundamental importance in the development and homeostasis ofmulticellular organisms4. Although autophagy is proposed to play arole in dying cells exposing ‘eat me’ signals48, our in vitro and in vivodata demonstrate that autophagy proteins also directly regulate thedegradation of apoptotic cells by targeting phagosome membranes inengulfing cells. Hownon-pathogen engulfments promote LC3 targetingremains to be determined, but it may relate in part to the mechanismof engulfment; as, interestingly, we found that phagocytosis of anuncoated latex bead does not trigger LC3 recruitment, consistentwith previous reports21,49, but differing from others50, whereas amacropinosome in the same cell does recruit LC3.Our data support a role for autophagy proteins downstream of the

ULK complex, and in particular LC3, in lysosome fusion51. Any suchroles for LC3 in autophagy would be obscured by the requirement ofLC3 and its homologues for autophagosome formation52,53. LC3 hasbeen argued to be dispensable for autophagosome–endosome fusionusing an in vitro reconstitution assay54, but such a role may exist inintact cells. Recently LC3B and the homologous protein GATE-16were shown to mediate membrane fusion between liposomes55,56.Although fusogenic activity may depend on assay conditions57, it isconceivable that LC3 mediates membrane fusion between phagosomesand lysosomes directly. Alternatively, one of the host of recentlyidentified LC3-interacting proteins could function in lysosome fusionand require LC3 for targeting58.Roles for entosis in tumour promotion, through the generation

of aneuploidy59, and suppression, based on the elimination ofmatrix-detached cells, have been proposed22. Matrix detachment isencountered by tumour cells at early stages of tumour formation,and later stages as cells metastasize in the body and are displacedfrom their niches22,60. One hallmark of tumour cells is the ability



ART I C L E S

Perc

enta

ge o

f LG

G1

-po

sitiv

e p

hag

oso

mes

Perc

enta

ge o

f R

AB

-5

-po

sitiv

e p

hag

oso

mes

DIC

mC

herr

y::

RA

B-5

GF

P::

LG

G-1

bec-1

RN

Ai

Co

ntr

ol R

NA

i ~320 min ~380 min ~420 min ~460 minBean stage Comma stage 1.5-fold stage

~320 min ~380 min ~420 min ~460 min

Control RNAi bec-1 RNAi

Bean Comma 1.5-fold 2-fold

Co

rpses p

er

em

bry

o

bec-1 RNAi Control RNAi

∗∗∗∗∗∗∗∗∗

∗∗∗

∗∗∗

∗∗∗

RAB-5

LGG-1

DIC

0:28:03 0:33:03 0:38:03 0:43:03 0:48:03 0:53:03

0

25

50

75

0

25

50

75

100

0

5

10

15

CED-1::GFP CED-1::GFP

a Control RNAi bec-1 RNAi

LGG-1

DIC

c d

e f

g

b

Control

RNAi

bec-1

RNAi

Control

RNAi

bec-1

RNAi

2-fold stage

n = 9 n = 16 n = 18 n = 14 n = 16 n = 10 n = 15n = 9

Figure 6 LGG-1 recruitment to apoptotic phagosomes during C. elegansembryonic development. (a) A C. elegans embryo expressing mCherry::RAB-5and GFP::LGG-1. The arrows point to two apoptotic phagosomes. SeeSupplementary Movie S11. Scale bar, 10 µm. (b) Cropped time-lapseimages showing RAB-5 and LGG-1 recruitment to an apoptotic phagosome(from a, left arrow). (c) Representative images of apoptotic phagosomesin embryos from control-RNAi- or bec-1-RNAi-fed worms. The insetsshow higher magnifications of the areas outlined in the main panels.(d) Quantification of GFP::LGG-1- and mCherry::RAB-5-positive phagosomes

in control- or bec-1-RNAi embryos determined by time-lapse imaging.Control, 21 embryos, 52 phagosomes; bec-1, 9 embryos, 25 phagosomes;∗∗∗P <0.0001, chi-squared test. (e) Representative DIC images from control-and bec-1-RNAi embryos at different stages of development (minutes afterfirst division). The yellow arrowheads mark apoptotic corpses. Scale bar,10 µm. (f) Quantification of apoptotic corpses in control- and bec-1-RNAiembryos; data show ± s.d.; ∗∗∗P < 0.0001. (g) Representative images ofCED-1::GFP embryos with control or bec-1 RNAi. The yellow arrowheadsmark engulfed corpses surrounded by CED-1::GFP. Scale bar, 5 µm.

to survive and proliferate in the absence of matrix adhesion, forexample in soft agar. Our data identify the mechanism by which entosiseliminates matrix-deprived cells. Entotic cells are killed by their hosts,in an autophagy protein-dependent manner. That autophagy proteininhibitors can significantly increase the level of transformed growth ofcells undergoing high rates of entosis, in amanner similar to Rho-kinaseinhibition, indicates that entosis may suppress transformed growth byinducing cell death, using machinery of the autophagy pathway. �

METHODSMethods and any associated references are available in the onlineversion of the paper at http://www.nature.com/naturecellbiology

Note: Supplementary Information is available on the Nature Cell Biology website

ACKNOWLEDGEMENTSWe thank J. Durgan, I. Ganley, X. Jiang, G. Mouneimne, E. Yao, A. Spencer andmembers of the Overholtzer laboratory for helpful discussions, reagents and forreading the manuscript. We also thank N. Lampen of the Memorial Sloan KetteringCancer Center Electron Microscopy Facility for processing of electron microscopysamples. This work was financially supported by a grant from the National CancerInstitute (CA154649;M.O.), TheGeoffrey BeeneCancer ResearchCenter atMSKCC(M.O.), the Louis V. Gerstner, Jr. Young Investigators Fund (M.O. and C.M.H.) andthe Alfred W. Bressler Scholar Fund (C.M.H.).

AUTHOR CONTRIBUTIONSO.F. and M.O. designed, carried out experiments and wrote the paper. S.E.K. andC.P.S. contributed experimental assistance and data. C.M.H. provided worm strainsand carried out RNAi feeding of C. elegans.

COMPETING FINANCIAL INTERESTSThe authors declare no competing financial interests.

Published online at http://www.nature.com/naturecellbiologyReprints and permissions information is available online at http://www.nature.com/reprints

1. Kumari, S., Mg, S. & Mayor, S. Endocytosis unplugged: multiple ways to enter thecell. Cell Res. 20, 256–275 (2010).

2. Kawane, K. et al. Requirement of DNase II for definitive erythropoiesis in the mousefetal liver. Science 292, 1546–1549 (2001).

3. Broadie, K. Axon pruning: an active role for glial cells. Curr. Biol. 14,R302–R304 (2004).

4. Elliott, M. R. & Ravichandran, K. S. Clearance of apoptotic cells: implications inhealth and disease. J. Cell Biol. 189, 1059–1070 (2010).

5. Savina, A. & Amigorena, S. Phagocytosis and antigen presentation in dendritic cells.Immunol. Rev. 219, 143–156 (2007).

6. Haas, A. The phagosome: compartment with a license to kill. Traffic 8,311–330 (2007).

7. Steinberg, B. E. & Grinstein, S. Pathogen destruction versus intracellular survival:the role of lipids as phagosomal fate determinants. J. Clin. Invest. 118,2002–2011 (2008).

8. Kinchen, J. M. et al. A pathway for phagosome maturation during engulfment ofapoptotic cells. Nat. Cell Biol. 10, 556–566 (2008).

9. Yang, Z. & Klionsky, D. J. Eaten alive: a history of macroautophagy. Nat. Cell Biol.12, 814–822 (2010).

10. Yang, Z. & Klionsky, D. J. Mammalian autophagy: core molecular machinery andsignaling regulation. Curr. Opin. Cell Biol. 22, 124–131 (2010).

11. Jin, S. & White, E. Tumor suppression by autophagy through the management ofmetabolic stress. Autophagy 4, 563–566 (2008).

12. Burman, C. & Ktistakis, N. T. Autophagosome formation in mammalian cells. Semin.Immunopathol. 32, 397–413 (2010).

13. Levine, B. & Yuan, J. Autophagy in cell death: an innocent convict? J. Clin. Invest.115, 2679–2688 (2005).

14. Rich, K. A., Burkett, C. & Webster, P. Cytoplasmic bacteria can be targets forautophagy. Cell Microbiol. 5, 455–468 (2003).

15. Yano, T. et al. Autophagic control of listeria through intracellular innate immunerecognition in Drosophila. Nat. Immunol. 9, 908–916 (2008).

16. Gutierrez, M. G. et al. Autophagy is a defense mechanism inhibiting BCGand Mycobacterium tuberculosis survival in infected macrophages. Cell 119,753–766 (2004).

17. Nakagawa, I. et al. Autophagy defends cells against invading group A Streptococcus.Science 306, 1037–1040 (2004).

18. Ogawa, M. et al. Escape of intracellular Shigella from autophagy. Science 307,727–731 (2005).

19. Ling, Y. M. et al. Vacuolar and plasma membrane stripping and autophagicelimination of Toxoplasma gondii in primed effector macrophages. J. Exp. Med. 203,2063–2071 (2006).

20. Sanjuan, M. A., Milasta, S. & Green, D. R. Toll-like receptor signaling in thelysosomal pathways. Immunol. Rev. 227, 203–220 (2009).



http://www.nature.com/naturecellbiology/

http://www.nature.com/naturecellbiology

http://www.nature.com/reprints

http://www.nature.com/reprints

ART I C L E S

21. Sanjuan, M. A. et al. Toll-like receptor signalling in macrophages links the autophagypathway to phagocytosis. Nature 450, 1253–1257 (2007).

22. Overholtzer, M. et al. A nonapoptotic cell death process, entosis, that occurs bycell-in-cell invasion. Cell 131, 966–979 (2007).

23. Overholtzer, M. & Brugge, J. S. The cell biology of cell-in-cell structures. Nat. Rev.Mol. Cell Biol. 9, 796–809 (2008).

24. Seglen, P. O. & Gordon, P. B. 3-methyladenine: specific inhibitor of au-tophagic/lysosomal protein degradation in isolated rat hepatocytes. Proc. Natl Acad.Sci. USA 79, 1889–1892 (1982).

25. Ichimura, Y. et al. A ubiquitin-like system mediates protein lipidation. Nature 408,488–492 (2000).

26. Hanada, T. et al. The Atg12-Atg5 conjugate has a novel E3-like activity for proteinlipidation in autophagy. J. Biol. Chem. 282, 37298–37302 (2007).

27. Petiot, A., Ogier-Denis, E., Blommaart, E. F., Meijer, A. J. & Codogno, P.Distinct classes of phosphatidylinositol 3’-kinases are involved in signalingpathways that control macroautophagy in HT-29 cells. J. Biol. Chem. 275,992–998 (2000).

28. Jung, C. H. et al. ULK-Atg13-FIP200 complexes mediate mTOR signaling to theautophagy machinery. Mol. Biol. Cell 20, 1992–2003 (2009).

29. Hosokawa, N. et al. Nutrient-dependent mTORC1 association with theULK1-Atg13-FIP200 complex required for autophagy. Mol. Biol. Cell 20,1981–1991 (2009).

30. Ganley, I. G. et al. ULK1.ATG13.FIP200 complex mediates mTOR signaling and isessential for autophagy. J. Biol. Chem. 284, 12297–12305 (2009).

31. Stenmark, H., Aasland, R., Toh, B. H. & D’Arrigo, A. Endosomal localization of theautoantigen EEA1 is mediated by a zinc-binding FYVE finger. J. Biol. Chem. 271,24048–24054 (1996).

32. Eskelinen, E. L. et al. Inhibition of autophagy in mitotic animal cells. Traffic 3,878–893 (2002).

33. Kimura, S., Noda, T. & Yoshimori, T. Dissection of the autophagosome maturationprocess by a novel reporter protein, tandem fluorescent-tagged LC3. Autophagy 3,452–460 (2007).

34. Sulston, J. E. & Horvitz, H. R. Post-embryonic cell lineages of the nematode,Caenorhabditis elegans. Dev. Biol. 56, 110–156 (1977).

35. Sulston, J. E., Schierenberg, E., White, J. G. & Thomson, J. N. The embryonic celllineage of the nematode Caenorhabditis elegans. Dev. Biol. 100, 64–119 (1983).

36. Ruck, A. et al. The Atg6/Vps30/Beclin1 ortholog BEC-1 mediates endocyticretrograde transport in addition to autophagy in C. elegans. Autophagy 7,386–400 (2011).

37. Zhou, Z., Hartwieg, E. & Horvitz, H. R. CED-1 is a transmembrane receptor thatmediates cell corpse engulfment in C. elegans. Cell 104, 43–56 (2001).

38. Kamada, Y., Sekito, T. & Ohsumi, Y. Autophagy in yeast: a TOR-mediated responseto nutrient starvation. Curr. Top. Microbiol. Immunol. 279, 73–84 (2004).

39. Kim, D. H. et al. mTOR interacts with raptor to form a nutrient-sensitive complexthat signals to the cell growth machinery. Cell 110, 163–175 (2002).

40. Mizushima, N. et al. A protein conjugation system essential for autophagy. Nature395, 395–398 (1998).

41. Mizushima, N., Sugita, H., Yoshimori, T. & Ohsumi, Y. A new protein conjugationsystem in human. The counterpart of the yeast Apg12p conjugation system essentialfor autophagy. J. Biol. Chem. 273, 33889–33892 (1998).

42. Hara, T. et al. FIP200, a ULK-interacting protein, is required for autophagosomeformation in mammalian cells. J. Cell Biol. 181, 497–510 (2008).

43. Cheong, H. et al. Atg17 regulates the magnitude of the autophagic response.Mol. Biol. Cell 16, 3438–3453 (2005).

44. Kabeya, Y. et al. Atg17 functions in cooperation with Atg1 and Atg13 in yeastautophagy. Mol. Biol. Cell 16, 2544–2553 (2005).

45. Kamada, Y. et al. Tor-mediated induction of autophagy via an Apg1 protein kinasecomplex. J. Cell Biol. 150, 1507–1513 (2000).

46. Zhao, Z. et al. Autophagosome-independent essential function for the autophagyprotein Atg5 in cellular immunity to intracellular pathogens. Cell Host Microbe 4,458–469 (2008).

47. Lee, H. K. et al. In vivo requirement for Atg5 in antigen presentation by dendriticcells. Immunity 32, 227–239 (2010).

48. Qu, X. et al. Autophagy gene-dependent clearance of apoptotic cells duringembryonic development. Cell 128, 931–946 (2007).

49. Huang, J. et al. Activation of antibacterial autophagy by NADPH oxidases. Proc. NatlAcad. Sci. USA 106, 6226–6231 (2009).

50. Shui, W. et al. Membrane proteomics of phagosomes suggests a connection toautophagy. Proc. Natl Acad. Sci. USA 105, 16952–16957 (2008).

51. Kabeya, Y. et al. LC3, a mammalian homologue of yeast Apg8p, is localized inautophagosome membranes after processing. EMBO J. 19, 5720–5728 (2000).

52. Weidberg, H. et al. LC3 and GATE-16/GABARAP subfamilies are both essential yetact differently in autophagosome biogenesis. EMBO J. 29, 1792–1802 (2010).

53. Fujita, N. et al. An Atg4B mutant hampers the lipidation of LC3 par-alogues and causes defects in autophagosome closure. Mol. Biol. Cell 19,4651–4659 (2008).

54. Morvan, J. et al. In vitro reconstitution of fusion between immature autophagosomesand endosomes. Autophagy 5, 676–689 (2009).

55. Nakatogawa, H., Ichimura, Y. & Ohsumi, Y. Atg8, a ubiquitin-like protein requiredfor autophagosome formation, mediates membrane tethering and hemifusion. Cell130, 165–178 (2007).

56. Weidberg, H. et al. LC3 and GATE-16 N termini mediate membrane fusion processesrequired for autophagosome biogenesis. Dev. Cell 20, 444–454.

57. Nair, U. et al. SNARE proteins are required for macroautophagy. Cell 146, 290–302.58. Behrends, C., Sowa, M. E., Gygi, S. P. & Harper, J. W. Network organization of the

human autophagy system. Nature 466, 68–76 (2010).59. Krajcovic, M. et al. A non-genetic route to aneuploidy in human cancers.

Nat. Cell Biol. 13, 324–330 (2011).60. Horbinski, C., Mojesky, C. & Kyprianou, N. Live free or die: tales of homeless (cells)

in cancer. Am. J. Pathol. 177, 1044–1052 (2010).



METHODS DOI: 10.1038/ncb2363

METHODSAntibodies and reagents. The following antibodies were used at the indicateddilutions: anti-LC3B (1:2,000) and anti-tubulin (1:2,000; Sigma), anti-ATG5(1:1,000), anti-ATG7 (1:1,000) and anti-phospho ribosomal protein S6 (1:1,000; CellSignaling) and anti-GAPDH (1:1,000; SCBT). The following inhibitors and reagentswere used at the indicated concentrations, 3-MA (Sigma) 3mM; Y27632 (EMD)10 µM and LysoTracker Red DND-99 (Invitrogen) 50 nM.

Constructs. pBabe–GFP–LC3, pBabe–mCherry–LC3, pMLP–IRES–shATG5 and2×FYVE–mCherry (FYVE domain from Hrs) were a gift from X. Jiang, MSKCC,USA. Tandem pBabe–GFP–mCherry–LC3 was a gift from J. Debnath, UCSF,USA. pRetro–Lamp1–GFP was cloned by inserting the rat Lamp1 sequence (fromaddgene plasmid#1817) into the BamHI–EcoRI sites of the pRetroQ–AcGFP1–N1vector (Clontech). pBabe–Cathepsin-B–mCherrywas cloned by insertingCathepsin-B–mCherry into the BamHI–SalI sites of pBabepuro. The pBabepuro–Bcl2 andpLXSN–HPV–E7 constructs were a gift from J. S. Brugge, Harvard MedicalSchool, USA, and have been described elsewhere61. The pBabe–H2B–mCherry andpBabe–mCherry–CAAX constructs were gifts from C. Leong and J. S. Brugge,Harvard Medical School, USA.

Virus production and infection. For virus infection, cells were seeded in a 6-wellplate at 5× 104 per well. The next day, 1ml viral supernatant was added with10 µgml−1 polybrene for 24 h followed by a media change. Cells were then selectedwith the appropriate antibiotic, puromycin (2 µgml−1), G418 (200 µgml−1) orblasticidin (10 µgml−1).

siRNA and shRNA. siGenome SMART pool siRNAs and individual oligonu-cleotides against human VPS34, ATG5, ATG7 and FIP200 were obtained fromDharmacon. MCF10A cells were seeded in a 6-well plate at 6× 104 per welland transfected with 100 nM siRNA using Oligofectamine (Invitrogen). Cells wereroutinely assayed 48 h post-transfection. The mouse Atg5 shRNA construct (DNAsequence 5′-CTGTTTACAGTCAGTCTAT-3′) in the retroviral MLP vector was agift fromX. Jiang,MSKCC,USA. Cell pools expressing theAtg5 shRNAwere assayedfollowing transduction and drug selection.

Tissue culture. MCF10A cells were cultured as previously described62 in DME/F12+5% horse serum, 20 ngml−1 EGF, 10 µgml−1 insulin or 100 ngml−1 IGF-I,0.5 µgml−1 hydrocortisone, 100 ngml−1 cholera toxin, 50Uml−1 penicillin andpenicillin/streptomycin. MCF10A cells overexpressing E7 and Bcl2 were generatedas described previously61. MCF7 cells were obtained from the Lombardi CancerCenter at Georgetown University; J774 mouse macrophages (ATCC) and HEK293cells (ATCC) were cultured in DMEM+ 10% FBS with penicillin/streptomycin.U937 human monocytic cells were cultured in nRPMI-1640 + 10% FBS withpenicillin/streptomycin.

Microscopy. To quantify entotic cell fate, 1.5× 105 cells were seeded overnighton 35mm glass-bottomed dishes (MatTek) and cell-in-cell structures were imagedby time-lapse microscopy the next day. Fluorescence and differential interferencecontrast (DIC) micrographs were acquired every 4min for 20 h using a Nikon Ti-Einverted microscope attached to a CoolSNAP CCD (charge-coupled device) camera(Photometrics), and NIS Elements software (Nikon). Only live internalized cellsat the start of time-lapse imaging were quantified for cell fate. Internalized cellswere scored for release, death or no change. Control experiments demonstratedthat >95% of cell-in-cell structures chosen for analysis were completely enclosedand shielded from external media at the start of imaging, as assessed by FM4-64staining. Entotic structures chosen by DIC were stained with FM4-64, which labelsmembranes in contact with external media. The numbers of structures stainingpositive or negative were quantified (Supplementary Fig. S3c). The timing andtype of internalized cell death was determined by DIC morphology, cessation ofmovement or, where present, monitoring nuclear markers. Inhibitors were added,as indicated, 30min before image acquisition. For mixed cultures, H2B–mCherry-expressing cells were plated with GFP–LC3-expressing cells at a 1:1 ratio. Confocalimaging was carried out with the Ultraview Vox spinning-disc confocal system(Perkin Elmer) equipped with a Yokogawa CSU-X1 spinning-disc head and anEMCCD camera (Hamamatsu C9100-13), and coupled to a Nikon Ti-Emicroscope.All imaging with live cells was carried out in incubation chambers at 37 ◦C and 5%CO2. Confocal image acquisition and analysis was carried out with Volocity software(Perkin Elmer).

Electron microscopy. Cells were grown on grid-etched Aclar film, and imagedby fluorescence microscopy until LC3 recruitment to vacuole membranes. Sampleswere then fixed in 2.5% glutaraldehyde/2%paraformaldehyde in 0.075M cacodylatebuffer at pH 7.5 for 1 h followed by rinsing in cacodylate buffer and post-fixation in2%osmium tetroxide for 1 h. The samples were then rinsed in double-distilled waterfollowed by dehydration in a graded series of alcohol (50%, 75%, 95% to absolute

alcohol) and overnight in 1:1 propylene oxide/Poly Bed 812. The following day, thesamples were embedded in Poly Bed 812 inside BEEM capsules and cured in an ovenat 60 ◦C for two days. Following immersion in liquid nitrogen, the Aclar film wasthen ripped from the capsules and ultrathin sections were obtained using a ReichertUltracut S microtome. Sections were stained with uranyl acetate and lead citrate.Images were obtained using a JEOL 1200 EX transmission electron microscope.

Measuring autophagy using tandem GFP–mCherry–LC3. MCF10A cellsexpressing GFP–mCherry–LC3 were plated on glass-bottom coverslip dishes.Internalized entotic cells, or control single cells, ±HBSS (12 h), were imagedby confocal microscopy in the presence or absence of VPS34, ATG5 or FIP200siRNA. Fluorescence intensities of mCherry and GFP were measured in cells usingVolocity software. Data were normalized against background and mCherry/GFPratios calculated.

Measuring autophagy by electron microscopy. MCF10A cells were seeded onnon-adherent plates for 6 h to induce a high frequency of cell-in-cell formation.Cells were then seeded onto glass coverslips overnight. Following this, cells weretrypsinized, pelleted and fixed in 2.5% glutaraldehyde/2% paraformaldehyde in0.075M cacodylate buffer at pH 7.5 for 1 h. Pellets were then processed as above.Three or more fields of view were acquired within the cytoplasm of internalized orsingle cells; ten individual cells were used in each group. The area of autophagosomesand autolysosomes was calculated as a percentage of cytoplasm using ImageJsoftware (NIH).

Apoptotic phagocytosis, latexbeadandmacropinosomeassays. For apoptoticphagocytosis assays, J774 cells were seeded onto glass-bottom coverslip dishes inthe presence of 200Uml−1 IFNγ for 2 days. U937 cells expressing H2B–mCherrywere stimulated with 80 Jm−2 to induce apoptosis and added to J774 cultures at4×106 per dish. Where indicated, 100 nM concanamycin A was added before imageacquisition. For latex beads (Polysciences), 3 µm beads were added to macrophagesat a ratio of 10:1. Cells were monitored for 20 h, and z stacks were acquired every10min. For MCF10Amacropinosome assays, red fluorescent dextran with a relativemolecular mass of 10,000 (Mr10K) was added to culture media (0.1mgml−1) for10min, followed by washing in media for 5min, and confocal imaging (0.5 µmz stacks) of red dextran-labelled macropinosomes at 1min intervals for 30minto monitor recruitment of GFP–LC3. For J774 macropinosome assays, cells wereimaged by confocal microscopy in the presence ofMr10K red dextran (0.1mgml−1).LysoTracker red (Invitrogen) was added to cultures (50 nm) after the time-lapseimaging to image acidification of phagosomes.

ImagingC. elegans apoptotic phagocytosis. ThemCherry::RAB-5; GFP::LGG-1-expressing strain was generated by crossing the pie-1pr ::mCherry::rab-5 allele fromstrain RT1043 into the lgg -1pr ::lgg -1::gfp-expressing strain (DA2123). Both strainswere provided by the Caenorhabditis Genetics Center (University of Minnesota,Minneapolis). RNAi experiments were carried out by bacterial feeding as describedpreviously63. Embryos obtained from dissected gravid hermaphrodites were placedon 2% agarose pads and mounted on a coverslip for observation using an UltraviewVox spinning-disc confocal system (Perkin Elmer) equipped with a YokogawaCSU-X1 spinning-disc head and an EMCCD camera (Hamamatsu C9100-13), andcoupled to a Nikon Ti-E microscope. For apoptotic corpse counting, embryos atdifferent developmental stages, as assessed by morphology, were imaged by DIC.Corpses were identified by morphology and counted.

Western blotting. Cells were scraped into ice-cold RIPA buffer (50mM Tris atpH 7.4, 150mM NaCl, 2mM EDTA, 1% NP40, 0.1% SDS plus protease inhibitorcocktail) and lysed for 10min on ice. Lysates were centrifuged at 15,870g at 4 ◦C for12min and protein was quantified by BCA assay (Pierce). Samples were separatedon 10% polyacrylamide SDS–PAGE gels (15% for LC3 blots), and transferred toa polyvinyldifluoride membrane. The membrane was blocked into TBS-T plus 5%BSA and incubated overnight at 4 ◦C with primary antibodies diluted in blockingbuffer. Blots were incubated with horseradish peroxidase conjugated to secondaryantibodies and protein was detected using enhanced chemiluminescence detection(Invitrogen). Densitometry analysis was carried out using ImageJ software (NIH).

Quantitative PCR. Total RNA was prepared using the RNAeasy mini kit(Qiagen) 48 h after transfection of cells with siRNA. Quantitative PCRwas carried out using the Bio-Rad iCycler real-time system (MyiQ), withSYBR green detection (iScript One-Step RT–PCR Kit with SYBR green(Bio-Rad)). Samples were analysed by the standard curve method in trip-licate. Reactions contained a single product as determined by agarose gelelectrophoresis and melting curve analysis. The following primer pairs wereused: ACTIN forward 5′-AGAGCTACGAGCTGCCTGAC-3′; ACTIN reverse5′-AGCACTGTGTTGGCGTACAG-3′; FIP200 forward 5′-GTGCTGGGACGGA-TACAAAT-3′; FIP200 reverse 5′-TTTCCAATGCAAGCTGTGTC-3′.

NATURE CELL BIOLOGY


DOI: 10.1038/ncb2363 METHODS

Cell-in-cell formation—suspension assay. MCF10A (1× 105 cells) pretreatedwith VPS34 or ATG5 siRNA, or in the presence of 3-MA or Y27632, were platedin non-adherent 6-well dishes for 7 h. Cytospins of cells were made and stainedfor E-cadherin and β-catenin. The percentage of cells in cell-in-cell structures wasquantified by counting >300 cells per slide.

Colony formation in soft agar. MCF10A cells (5× 104) overexpressing E7 andBcl2 were added to 1ml of growth medium with 0.3% agar and layered onto1ml of 0.5% agar beds in 6-well plates and covered in 1ml liquid growth media.Cells were fed with 1ml of liquid medium every 5 days for 2 weeks, afterwhich colonies were stained with 0.02% iodonitrotetrazolium chloride (Sigma-Aldrich) and quantified using an Optronix Gelcount colony counter (OxfordOptronix). Assays were conducted in duplicate in three independent experiments.Where indicated, 3-MA (3mM) or Y27632 were included in all solutions. Counts

of cell-in-cell structures were made by light microscopy, 48 h after seeding insoft agar.

Statistics. Indicated P values were obtained using Student’s t -test, or the chi-squared test where indicated.

61. Debnath, J. et al. The role of apoptosis in creating and maintaining luminalspace within normal and oncogene-expressing mammary acini. Cell 111,29–40 (2002).

62. Debnath, J., Muthuswamy, S. K. & Brugge, J. S. Morphogenesis and oncogenesisof MCF-10A mammary epithelial acini grown in three-dimensional basementmembrane cultures. Methods 30, 256–268 (2003).

63. Kamath, R. S., Martinez-Campos, M., Zipperlen, P., Fraser, A. G. & Ahringer,J. Effectiveness of specific RNA-mediated interference through ingested double-stranded RNA in Caenorhabditis elegans. Genome Biol. 2, 1–10 (2000).

NATURE CELL BIOLOGY


S U P P L E M E N TA RY I N F O R M AT I O N

WWW.NATURE.COM/NATURECELLBIOLOGY 1

DOI: 10.1038/ncb2363

Figure S1 LC3 recruitment to entotic vacuoles. (a) Images of HEK293 and MCF7 cell-in-cell structures expressing tandem GFP-mCherry-LC3; only GFP is shown. Arrows mark LC3 recruitment. Bar = 15mm. (b) Measurements of GFP-LC3 fluorescence during early recruitment to entotic vacuoles in two MCF10A cell-in-cell structures. Fluorescence was measured at both the vacuole membrane and the cytoplasm (Cyto), and is presented as fold difference over time zero. Individual data points are mean +/-SD intensity of GFP fluorescence, of 3 separate >8uM2 areas of vacuole

membrane or cytoplasm from a confocal image taken through the middle of the cell at each timepoint. Fold changes in autophagosome number (Puncta) were monitored during LC3 recruitment, by counting LC3 puncta through each plane of a live confocal z-series taken at 0.5mm intervals covering the entire cell. Linear regression analysis shows no increase in LC3 puncta during LC3 recruitment to the entotic vacuole. (c) Wider field images, DIC and EM, of the cell-in-cell structure used for correlative electron microscopy in Figure 1d.

0 5 10 15 20

0.5

1.0

1.5

2.0

0 5 10 15 20

0.5

1.0

1.5

2.0

Time (minutes) Time (minutes)

aFl

uore

scen

ce F

old

Cha

nge

Fluo

resc

ence

Fol

d C

hang

e

b

DIC EM

Supplementary Figure S1

c

Vacuole 1Vacuole 2

Puncta 1 Puncta 2

Cyto 1Cyto 2



2 WWW.NATURE.COM/NATURECELLBIOLOGY

Figure S2 Autophagy inhibition. (a) Percentage of MCF10A cell-in-cell structures, with and without 3mM 3-MA, that show LC3 recruitment associated with death of internalized cell. Data represent mean +/- SEM from at least 3 independent experiments, n=total number of death events. *P<0.03, (b) Entotic cell fate of cells treated with 3mM 3-MA, +/-SEM #P<0.02, ##P<0.007, from 3 independent experiments, n= total number of structures analyzed. Lysates from MCF10A cells treated with siRNA against

(c) VPS34, (d) ATG5 and (e) ATG7 were probed for protein knockdown by western blotting. (f) Levels of FIP200 mRNA were detected by qRT-PCR following siRNA treatment. Graph shows analysis in triplicate, error bars equal SD. **P<0.001. (g) Autophagy inhibition was measured by LC3 western blot from MCF10A cells treated with siRNA against VPS34, ATG5, ATG7, and FIP200 or treated with the inhibitor 3-MA (3mM). All reagents inhibited formation of LC3-II after 6 hours starvation in HBSS.

ATG5siControl FIP200siControl- - - -+ + + +

3-MAControl- -+ +Starved

WB: LC3

ATG

5si

Con

trol

ATG5

GAPDH

ATG

7si

Con

trol

Tubulin

bc e

LC3II/LC3Iratio 0.35 0.58 0.46 0.360.69 1.18 0.5 0.63 0.19 0.74 0.27 0.18

LC3ILC3II

0.0

0.4

0.8

1.2

FIP

200/

actin

**

Control FIP200si

ATG7siControl- -+ +

0.21 0.45 0.25 0.13

d

ATG7V

ps34

si

Con

trol

Vps34

GAPDH

Vps34siControl- -+ +

0.37 0.6 0.36 0.41


n=159 n=170

#

##Control 3-MA

0

25

50

75

100

Cel

l Fat

e %

f

Control 3-MA0

25

50

75

100

n=68 n=37

LC

3 R

ing

%

*

a

g




Figure S3 Internalized entotic cells have restricted access to external growth factors. (a) Entotic structures of MCF10A cells expressing mCherry-CAAX (plasma membrane marker) were incubated in the presence of fluorescent EGF-Alexa488 and imaged by confocal timelapse microscopy. Bar = 10mm. (b) MCF10A cell-in-cell structures were formed and then incubated in HBSS (starved) or full media for 16 hours. Images are representative confocal slices after immuno-staining with anti-Phospho

S6 Ribosomal Protein and DAPI. Asterisk marks reduced Phospho S6 Rp staining of internalized cell. Bar = 10mm. (c) Quantification of entotic structures where entotic membrane is stained negative (therefore fully internalized) or positive (not fully internalized), n=97 structures. Representative images of cell-in-cell structures showing FM4-64 negative (internalized) and positive (not fully internalized) entotic vacuole membranes.

Full Media Starved

DA

PIPh

osph

o S6

Rp

*

0:00:00

0:59:58

7:20:02

mCherry-CAAX / EGF-Alexa488 DIC


a

b

FM4-64

FM4-64

FM4-

64 -v

e FM

4-64

+ve

FM4-64 -ve FM4-64 +ve0

20

40

60

80

100

% p

erce

ntag

e

c

n=97




Figure S4 LC3 recruits to phagosomes harboring apoptotic cells. (a) Control J774 and J774 shATG5 cell lysates were probed for Atg5 and Gapdh by western blotting. (b) Control J774 and J774 shATG5 cells expressing GFP-LC3 were imaged by confocal microscopy in full growth media or after 1 hr starvation in HBSS. Images are maximal projections of confocal z-stacks. (c) J774 cells were incubated with apoptotic U937 cells expressing H2B-mCherry in the presence of 100nM Concanamycin A

and imaged by confocal microscopy. Images show an engulfed apoptotic corpse with no H2B-mCherry diffusion over 14 hours. Final image is taken following FM4-64 staining, where the phagosome is negative for FM4-64, demonstrating complete engulfment. Images are representative of 9 cells. Bar = 10mm. (d) An additional example of GFP::LGG-1 and mCherry::Rab-5 recruitment to an apoptotic phagosome in a C.elegans embryo.


Full Media Starved

J774

GFP

-LC

3J7

74 G

FP-L

C3

shAT

G5

b

ATG5

GAPDH

J774

GFP

-LC3

J774

GFP

-LC3

shAT

G5

aD

ICR

ab-5

LGG

-1

d

DIC

H2B-mCherry FM4-64

DIC 00:00 02:40 05:00 14:00

c

+ C

onca

nam

ycin

A 1

00nM

14:00




Figure S5 J774 apoptotic corpse degradation. Control J774 GFP-LC3 (i-xxi) or J774 shATG5 GFP-LC3 (xxii-xxxxii) cells were incubated with apoptotic U937 cells expressing nuclear marker H2B-mCherry and imaged by spinning disk confocal microscopy every 10 minutes. Following

engulfment, degradation of apoptotic corpses was monitored by observing diffusion of H2B-mCherry signal. 80 minutes post engulfment, only control and shATG5 cells marked with a tick displayed complete mCherry diffusion.


020

8060

40Tim

e (mins)

ivi v iv iii ii

vii

xii xi x ix

viii

xiii

J774 GFP-LC3

xvi

xv

xivxxi

xx

ixx

xviii

xvii

xxii

020

8060

40Tim

e (mins)

xxiii

xxviii

xxvii

xxvi

xxv

xxiv

J774 shATG5 GFP-LC3

ixxx

xxxii

xxxi

xxx

xxxiii

xxxiv

xxxvii

xxxvi

xxxv

ixxxx

xxxviii

xxxxii

xxxxi

xxxx

x

020

8060

40Tim

e (mins)

020

8060

40Tim

e (mins)

020

8060

40Tim

e (mins)

02 0

8060

40Tim

e (mins)




Figure S6 Full scan blots. Molecular weights are indicated in kDa. Boxes show cropped images used in figures.


56

9572

56

9572

17

34 26

43

17

34 26

43

17

34 26

17

34 26

43

17

34 26

56

9572

26

43

34

4334

43

34

26

4334

Figure S2c

17

34 26

43

1309572

Figure S2d Figure S2e

Figure S2g Vps34siRNA ATG5siRNA ATG7siRNA

3MAFIP200siRNA

Figure S4a




Supplemental Movie Legends

Movie S1 LC3 recruitment to an entotic vacuole. An entotic cell-in-cell structure of MCF10A cells expressing GFP-LC3 and H2B-mCherry was imaged every 6 minutes by confocal microscopy. GFP-LC3 recruits to the entotic vacuole prior to non-apoptotic death of the internalized cell.

Movie S2 High-temporal analysis of GFP-LC3 during recruitment to an entotic vacuole. Movie shows a single confocal slice imaged every 13 seconds during GFP-LC3 recruitment, and every 5 minutes after. Inset shows corresponding DIC image.

Movie S3 Recruitment of LC3 associated with correlative electron microscopy. A cell-in-cell structure of MCF10A cells expressing GFP-LC3 was monitored by DIC and fluorescence every 3 minutes using time-lapse microscopy up to the point of LC3 recruitment to the entotic vacuole membrane. Samples were fixed and re-imaged before processing for correlative electron microscopy (see Figure 1d).

Movie S4 Timing of PI(3)P and LC3 recruitment to entotic vacuoles. Representative movie showing the timing between 2xFYVE-mCherry and GFP-LC3 recruitment to an entotic vacuole in MCF10A cells, imaged every 3 minutes by time-lapse microscopy.

Movie S5 Timing of LC3 and Lamp1 recruitment to entotic vacuoles. Representative movie showing timing between mCherry-LC3 and Lamp1-GFP recruitment to an entotic vacuole in MCF10A cells, imaged every 3 minutes by confocal microscopy.

Movie S6 Representative cell fates of entotic cells grown in soft agar. MCF10 E7+Bcl2 cells grown in soft agar. Left frame shows an entotic death event. Right frame shows a release event of entotic cells expressing the nuclear marker H2B-mCherry. Movie S7 LC3 recruits to entotic vacuoles harboring apoptotic inner cells. Apoptotic death of an internalized cell in an entotic structure of MCF10A cells expressing GFP-LC3 and H2B-mCherry. Note LC3 recruitment to vacuole after internalized cell apoptosis, as determined by H2B condensation and fragmentation. Cells were imaged every 4 minutes by time-lapse microscopy.

Movie S8 LC3 recruits to apoptotic phagsosomes in macrophages. J774 macrophages expressing GFP-LC3 phagocytizing apoptotic U937 cells expressing H2B-mCherry. Cells were imaged every 10 minutes by confocal microscopy.

Movie S9 LC3 recruits to macropinosomes. Macropinocytosis was monitored in J774 cells expressing GFP-LC3 incubated in red dextran and imaged every 3 minutes by confocal microscopy.

Movie S10 LC3 does not recruit to phagosomes housing uncoated latex beads. Macropinocytosis and phagocytosis of uncoated latex beads in J774 GFP-LC3 cells was monitored through time by confocal microscopy. Lysotracker red was added to media at the end of the experiment to label acidified compartments and to confirm complete engulfment of latex beads. Arrow marks a macropinosomes recruiting LC3, arrow heads mark engulfed latex beads.

Movie S11 LGG-1 recruits to apoptotic phagosomes during C. elegans embryonic development. C. elegans embryo expressing mCherry::Rab-5 and LGG-1::GFP. Note the recruitment of both markers to two apoptotic phagocytic events (arrows) at the left side of the embryo. Images captured every 5 minutes.


Documents

Autophagy machinery mediates macroendocytic processing and entotic cell death ... - Weebly · 2021. 2. 1. · DIC mCherryÐLC3 GFPÐLC3 G120A Merge 0 150 300 X X X X X Time (min)