Article

The HEAT repeat protein HPO-27 is a

lysosome fission factor

https://doi.org/10.1038/s41586-024-07249-8 Letao Li1,2,6, Xilu Liu1,3,6, Shanshan Yang1,3,6, Meijiao Li2,4, Yanwei Wu1, Siqi Hu1,3, Wenjuan Wang1,
Amin Jiang1, Qianqian Zhang1, Junbing Zhang1, Xiaoli Ma1, Junyan Hu1, Qiaohong Zhao2,
Received: 3 January 2023
Yubing Liu1, Dong Li1,3, Junjie Hu1,3, Chonglin Yang2,4, Wei Feng1,3 ✉ & Xiaochen Wang1,3,4,5 ✉
Accepted: 28 February 2024

Published online: xx xx xxxx

Lysosomes are degradation and signalling centres crucial for homeostasis,
Check for updates development and ageing1. To meet diverse cellular demands, lysosomes remodel
their morphology and function through constant fusion and fission2,3. Little is known
about the molecular basis of fission. Here we identify HPO-27, a conserved HEAT repeat
protein, as a lysosome scission factor in Caenorhabditis elegans. Loss of HPO-27 impairs
lysosome fission and leads to an excessive tubular network that ultimately collapses.
HPO-27 and its human homologue MROH1 are recruited to lysosomes by RAB-7 and
enriched at scission sites. Super-resolution imaging, negative-staining electron
microscopy and in vitro reconstitution assays reveal that HPO-27 and MROH1 self-
assemble to mediate the constriction and scission of lysosomal tubules in worms and
mammalian cells, respectively, and assemble to sever supported membrane tubes
in vitro. Loss of HPO-27 affects lysosomal morphology, integrity and degradation
activity, which impairs animal development and longevity. Thus, HPO-27 and MROH1
act as self-assembling scission factors to maintain lysosomal homeostasis and function.

Lysosomes are the major degradative organelles and signalling centres lipophagy in hepatocytes20. However, scission molecules that act on
that coordinate cellular metabolism with clearance1. Lysosomal dys- lysosomes remain to be identified. Huntingtin, elongation factor 3,
function is associated with metabolic disorders, neurodegenerative a subunit of protein phosphatase 2A, signalling kinase TOR1 (HEAT)
diseases and age-related pathologies, which highlights the importance repeats are arrays of amphiphilic α-helix pairs connected by a short
of lysosomes in physiology and ageing1,4. loop. They are found in functionally diverse eukaryotic proteins21. The
Lysosomes undergo constant fusion and fission. They fuse with other HEAT motifs are arranged in an antiparallel manner and stacked to form
organelles to acquire essential components and acidity, receive car- a structurally flexible and highly elastic solenoid with a large surface
goes for degradation and fulfil secretory functions2. To balance this area21,22. These properties are important for shaping scaffolds and
inward flow of materials, lysosomes undergo fission to maintain their bridging protein–protein interactions in various intracellular con-
steady-state number, shape, size, composition and function, and to texts21,22.
accomplish regeneration3. Soluble N-ethylmaleimide-sensitive fac- Here we identify the HEAT repeat protein HPO-27 as a lysosome scis-
tor attachment protein receptor (SNARE) complexes are assembled sion factor. HPO-27 and its human homologue MROH1 are recruited
in trans to drive the fusion of lysosomes with cargo delivery vesicles to lysosomes by RAB-7 to mediate the scission of lysosomal tubules.
after tethering by RAB-7 and/or ARL8 effectors2,5. Much less is known Loss of HPO-27 in C. elegans affects lysosomal morphology, structural
about lysosome membrane fission. The vesicle-forming machinery integrity and degradation activity, which in turn causes developmental
can assemble on purified dense lysosomes to initiate budding, medi- defects and reduced longevity.
ate phagolysosome resolution and contribute to the formation and
fission of tubules generated by autolysosome reformation (ALR)6–8.
The hereditary spastic paraplegia proteins SPG11, SPG15 and AP5 are Loss of hpo-27 affects lysosomes
implicated in lysosomal membrane recycling9–11. Phosphatidylinositol Lysosomes have both vesicular and tubular morphologies (Fig. 1a–c).
3-phosphate (PtdIns3P), phosphatidylinositol 4-phosphate (PtdIns4P), Abundant tubular lysosomes appear in C. elegans at moult during larval
phosphatidylinositol 4,5-bisphosphate (PtdIns(4,5)P2) and phos- development and during ageing, but they exhibit different proper-
phatidylinositol 3,4,5-trisphosphate (PtdIns(3,5)P2) are involved in ties23,24. In genetic screens for mutants with altered lysosome tubula-
the exit of membranes from lysosomes through vesicular or tubular tion, we isolated eight recessive mutations that cause hyper-tubulation
intermediates, but their downstream effectors are unclear8,12–17. ARF– of lysosomes, which was visualized by labelling with the lysosomal
PI4KIIIβ-positive vesicles may regulate PtdIns3P signalling to facilitate DNase II NUC-1 (NUC-1::Cherry) (Extended Data Fig. 1a,c,e,g,i,k,m,o).
lysosomal tubule fission18. Dynamin superfamily proteins mediate All eight mutations are in the gene hpo-27, which encodes a conserved
scission in many contexts19, and dynamin 2 severs ALR tubules during HEAT repeat protein homologous to human MROH1 (71% sequence
1
National Laboratory of Biomacromolecules, CAS Center for Excellence in Biomacromolecules, Institute of Biophysics, Chinese Academy of Sciences, Beijing, China. 2State Key Laboratory of
Conservation and Utilization of Bio-Resources in Yunnan, and Center for Life Sciences, School of Life Sciences, Yunnan University, Kunming, China. 3College of Life Sciences, University of
Chinese Academy of Sciences, Beijing, China. 4Southwest United Graduate School, Kunming, China. 5School of Life Sciences, Southern University of Science and Technology, Shenzhen,
Guangdong, China. 6These authors contributed equally: Letao Li, Xilu Liu, Shanshan Yang. ✉e-mail: wfeng@ibp.ac.cn; wangxiaochen@ibp.ac.cn

Nature | www.nature.com | 1
Article
Embryo L4 Adult day 1 g WT hpo-27(tm5336) n Dense vesicle o Tubule p Membrane whorl
a b c 100

P < 0.0001
No. of vesicular
80

per unit area

P < 0.0001
lysosomes

P < 0.0001
60
WT

40
20
0
Two-fold L4 Day 1
d e f h
embryo
q r s
hpo-27(tm5336)

WT hpo-27(tm5336) Dense vesicle Tubule Membrane whorl

tubules per unit area

80

No. of lysosomal
P < 0.0001
60
40 P < 0.0001
P < 0.0001
20
WT hpo-27(tm5336) 0
Two-fold L4 Day 1
j k l embryo
hyp7 hyp7 hyp7 i WT hpo-27(tm5336) t Tubule
40

lysosomal tubules
per unit area (μm)
P < 0.0001

Total length of
30 P < 0.0001
Seam cell Seam cell Seam cell
20
P = 0.0008
10
100% 42% 58%
0
Two-fold L4 Day 1
m v
P < 0.0001
P < 0.0001

WT WT embryo
w u
P < 0.0001

hpo-27(tm5336) Tubule
hpo-27(tm5336)
P = 0.0008
120
P < 0.0001

P = 0.0038

Lysosomal pattern in
NUC-1::Cherry in the
Worms with reduced

100 100
100

hpo-27(tm5336)
Lysosomes within
hypodermis (%)

each group (%)

mutants (%)
80 80 80
P < 0.0001

60
P < 0.0001

60
P = 0.2715

60
40 40 40
0 0 0 0 0 Dense vesicle
20 20 20 Tubule
Membrane whorl
0 0 0
L4 1 2 3 4 5 Dense Tubule Membrane 1 2 3
vesicle whorl Adult stage (day)
Adult stage (day)

Fig. 1 | Loss of HPO-27 affects lysosomal morphology. a–f,j–l, Confocal of worms with reduced NUC-1::Cherry signal (n = 100 worms in each of 3
fluorescence images of embryos (a,d), L4 larvae (b,e) and adult hypodermis independent experiments). n–u, TEM images of lysosomes in hypodermis of
(c,f,j–l) in WT (a–c,j) and hpo-27(tm5336) (d–f,k,l) worms expressing the WT (n–p) and hpo-27(tm5336) (q–u) adult day 1 animals. Yellow and red
lysosomal marker NUC-1::Cherry without (a–f) and with the seam cell marker arrowheads indicate tubules and membrane whorl-containing vesicles,
AJM-1::GFP ( j–l; adult day 2, n = 100). White arrowheads and arrows indicate respectively. v, Quantifications of n–u (n = 5 worms per strain). w, Lysosomal
vesicular and short tubular lysosomes, respectively, in WT. Blue arrowheads patterns in hpo-27(tm5336) mutants (left to right: n = 5, 5 and 7 worms). Data
indicate long tubules in hpo-27(tm5336) mutants. NUC-1::Cherry fluorescence show mean ± s.d. (g–i,m) or mean ± s.e.m. (v,w). Mutant and WT data were
is reduced in hyp7 cells but remains in the AJM-1::GFP-positive lateral seam cells compared by two-way analysis of variance (ANOVA) with Bonferroni post-test.
in hpo-27(tm5336) mutants. g–i, Numbers of vesicular (g) and tubular (h) P values are indicated. Scale bars, 0.5 μm (n–u) or 5 μm (a–f,j–l).
lysosomes and total length of lysosomal (i) tubules (n = 14 worms). m, Percentage

similarity and 23% sequence identity; Extended Data Figs. 1w,x and SCAV-3::tagRFPt and LMP-1::RFP to further analyse this phenotype26,27.
2a). HPO-27 and MROH1 are large helical proteins that contain 37 HEAT These lysosomal proteins exhibited a vesicular pattern in WT ani-
repeats with no other discernible domains (Extended Data Figs. 1x mals but stained extensive tubules in 18–44% of hpo-27 mutants at
and 2a). Seven hpo-27 alleles carried nonsense mutations (creating a adult day 2 (Extended Data Fig. 3i,j,l,m,o,p). In the rest of the hpo-27
premature stop codon in the coding region), whereas qx437 carried a worms, Cherry or RFP was barely seen in hyp7 cells (Extended Data
missense mutation (G316E; Extended Data Fig. 1x). The tubular lyso- Fig. 3k,n,q). The proportion of worms with reduced NUC-1::Cherry
some phenotype was also seen in tm5336, a hpo-27 deletion mutant, signals increased with age. At adult day 5, almost all hpo-27(tm5336)
or when hpo-27 was inactivated using RNAi (Fig. 1d–f,k and Extended worms lacked NUC-1::Cherry in hyp7 cells (Fig. 1m). After time-lapse
Data Fig. 1q,s–u″). imaging of hundreds of hpo-27(tm5336) adults, we captured the sudden
During development, wild-type (WT) C. elegans contains mainly vesic- collapse of the hyp7 lysosomal tubular network, which caused diffu-
ular lysosomes with a few short tubules, as indicated by NUC-1::Cherry sion of RNST-2::Cherry and SCAV-3::GFP in the cytosol of hyp7 cells
or the lysosomal amino acid transporter LAAT-1::GFP25 (Fig. 1a–c,j and (Extended Data Fig. 3r–v′ and Supplementary Video 1).
Extended Data Fig. 3a,c). In hpo-27(tm5336) mutants, abundant lyso- We next examined lysosomal structure by transmission electron
somal tubules were seen in multiple tissues at various developmen- microscopy (TEM). In WT day 1 adults, hypodermal lysosomes appeared
tal stages (Fig. 1d–f,k and Extended Data Figs. 1u–u″ and 3b,d). The as membrane-enclosed, dense, spherical vesicles (dense vesicle, 65%)
number of vesicular lysosomes was substantially reduced, whereas and electron-lucent tubules that contained a few parallel membrane
the abundance and length of lysosomal tubules, and the total lyso- layers (tubule, 30%) (Fig. 1n,o,v). In hpo-27(tm5336) mutants, the pro-
some volume, were strongly increased in hpo-27 mutants (Fig. 1g–i and portion of dense vesicular lysosomes was only 3.8%, whereas lysosomal
Extended Data Fig. 3e–h). hpo-27 mutant adults showed much reduced tubules were substantially increased (Fig. 1q,t–v). In hpo-27 mutants,
NUC-1::Cherry fluorescence in the hypodermis (Fig. 1l and Extended 38% of the hypodermal lysosomes were spherical vesicles that contained
Data Fig. 1b,d,f,h,j,l,n,p,r,v–v″). In these worms, NUC-1::Cherry was membrane whorls, which were rare in WT animals (Fig. 1p,r,v). The mem-
seen in the lateral seam cells labelled by AJM-1::GFP but became invis- brane whorl lysosomes were often localized close to or were connected
ible in the hyp7 cell, a multinuclear epidermal syncytium that extends to tubules (Fig. 1s). Both tubular and whorl-type structures were stained
throughout most of the animal, except for a few faint Cherry-positive by diaminobenzidine (DAB) in worms expressing SCAV-3::APEX, which
dots. Expression of a C. elegans fosmid covering the entire hpo-27 gene confirmed that they are lysosomal structures (Extended Data Fig. 4a–i).
efficiently rescued the reduced NUC-1::Cherry phenotype in hpo-27 The proportion of membrane whorl-containing vesicles increased
mutants (Extended Data Fig. 2b). We expressed the lysosomal endori- with age in hpo-27 mutants, whereas lysosomal tubules were reduced
bonuclease RNST-2::Cherry and the lysosomal membrane proteins (Fig. 1w). In hpo-27 mutants at adult day 3, when >80% worms showed

2 | Nature | www.nature.com
reduced NUC-1::Cherry fluorescence, few hypodermal lysosomes were
observed by TEM. These lysosomes contained membrane whorls or
cytosol content and seemed to be enclosed by ribosome-decorated
endoplasmic reticulum (ER) membranes (Extended Data Fig. 4j–n).
Lysosome density (lysosome area per μm2 of the epidermal surface)
was substantially reduced with age in hpo-27 mutants (Extended Data
Fig. 4o). Thus, loss of HPO-27 affects lysosomal morphology and integ-
rity. In hpo-27 mutants, excessive lysosomal tubules form networks,
which rupture to release their content.
In hpo-27 mutants, the ER network was disrupted, and mitochondria
were fragmentated and dilated (Extended Data Fig. 4p–s,v–y,ab–ae).
These defects appeared at adult day 3 and correlated well with lysosome
collapse. The expression of HPO-27 rescued the lysosome collapse
and abnormal ER and mitochondrial phenotypes in hpo-27 mutants
(Extended Data Figs. 2b and 4t,u,z,aa). An endogenous HPO-27::mKate2
reporter, created using CRISPR–Cas9, was not present on mitochondria
or the ER (Extended Data Fig. 4af–ah). Therefore, lysosomal rupture
may affect the ER and mitochondria.
Affinity-purified lysosomes from hpo-27(tm5336) worms exhibited
lower cathepsin B and cathepsin L activity than WT worms (Extended
Data Fig. 5a). Cleavage of NUC-1::Cherry, which is mediated by cath-
epsins after lysosomal delivery of the fusion protein23, was reduced
in hpo-27 mutants (Extended Data Fig. 5b,c). LGG-1 is the C. elegans
homologue of ATG8 and LC3. GFP::LGG-1 was removed through the
autophagy–lysosome pathway in WT worms but accumulated exten-
sively in autophagy-defective mutants such as epg-5(tm3425)28,29
(Extended Data Fig. 5d–m,s). hpo-27 mutant embryos, larvae and
adults contained significantly more GFP::LGG-1 puncta than WT ani-
mals (Extended Data Fig. 5n–s). Embryonic and germ cell corpses
were cleared by a lysosome-dependent pathway in WT worms but
persisted in hpo-27 mutants (Extended Data Fig. 5t–y). Loss of hpo-27
affected phagosome acidification and impaired cell corpse degrada-
tion in phagolysosomes, as measured by HIS-24::GFP loss (which labels
chromatin in all somatic nuclei, including cell corpses) (Extended Data
Fig. 6a–e). Thus, loss of hpo-27 affects lysosome activity, which impairs
the degradation of autophagic and phagocytic cargoes.
HPO-27 promotes lysosomal membrane fission
Endogenous HPO-27::mKate2 was expressed in embryonic, larval and
adult stages in multiple tissues (Extended Data Fig. 7a–j″). Expression of
HPO-27::GFP in hypodermal cells (driven by the semo-1 promoter), but
not the intestine (driven by the vha-6 promoter), rescued the lysosome
collapse phenotype in hyp7 cells (Extended Data Fig. 2b). This result
suggests that HPO-27 acts cell-autonomously to regulate lysosomes.
Expression of human MROH1 in hypodermis partially rescued the lyso-
some collapse phenotype in hpo-27(tm5336) mutants (Extended Data
Fig. 2b).
HPO-27::mKate2 and HPO-27::GFP exhibited a vesicular localiza-
tion pattern and associated with lysosomes labelled by NUC-1, SCAV-3
or RNST-2 in WT worms, and in cup-5 and ppk-3 mutants that contain
substantially enlarged lysosomes30,31 (Fig. 2a–c and Extended Data
Fig. 7k–k‴). In all three strains, HPO-27 was concentrated at specific foci
on the surface of about 60% of lysosomes (Fig. 2a–d and Extended Data
Fig. 7k–k‴). Next, we monitored lysosome dynamics in WT worms and
hpo-27 mutants. New tubules extended from vesicular or tubular lys-
osomes then underwent fission, retraction or fusion with other tubules
or vesicular lysosomes (Fig. 2e–h and Supplementary Video 2). Among
150 newly formed lysosomal tubules in WT worms, 51% were severed,
34% were retracted and 15% fused with other lysosomes (Fig. 2i). In
hpo-27(tm5336) mutants, fission was reduced to 27%, and the percent-
age of overall fusion events was increased. Among these events, fusions
between tubules were increased, whereas vesicle–tubule fusions were
reduced (Fig. 2i). Overexpression of HPO-27::GFP in WT animals reduced
the abundance and length of lysosomal tubules at moult and in day 3
adults (Fig. 2j–o). We analysed HPO-27::mKate2 dynamics at moult, a
stage when the hypodermis contains abundant and highly dynamic
lysosomal tubules23. Super-resolution time-lapse imaging showed
enrichment of HPO-27 in discrete foci at the neck or along the length
of the lysosomal tubule, and constriction and scission of the tubule
then occurred underneath or between the foci (Fig. 3a and Supplemen-
tary Videos 3 and 4). Figure 3a shows that the single HPO-27 punctum
splits into two when scission occurs, and one of the new puncta stayed
attached to the lysosomal neck and the other was lost. Thus, HPO-27
associates with lysosomes and enriches at scission sites to promote
tubule severing. Loss of HPO-27 impairs lysosomal fission, which causes
excessive tubulation and network formation, whereas overexpression
of HPO-27 may facilitate membrane scission, which leads to fewer,
shorter lysosomal tubules.
We tested whether dynamin is involved in lysosomal fission. Loss of
the C. elegans dynamin DYN-1 slightly increased the abundance and
length of lysosomal tubules in the hypodermis. The phenotypes were
much weaker than in hpo-27 mutants (Extended Data Fig. 6f–h,j,k). The
dyn-1(ky51) mutation did not alter the lysosomal tubule phenotype
in hpo-27 mutants or the lysosomal association of HPO-27 (Extended
Data Fig. 6i–n). DYN-1::GFP puncta were seen mainly at the apical
surface in the hypodermis and were not associated with lysosomes
(Extended Data Fig. 6o,p). These data suggest that DYN-1 does not
play a major part in hypodermal lysosome fission. Overexpression
of HPO-27 suppressed lysosomal tubules in WT animals and in dyn-1
mutants (Fig. 2j–o and Extended Data Fig. 6j). Whereas overexpres-
sion of DYN-1(WT) caused a reduction in lysosomal tubules in worms
treated with control RNAi and hpo-27-targeted RNAi, overexpression
of DYN-1(T67A), a putative GTP-bound constitutively active dynamin32,
reduced lysosomal tubules to a much lesser extent than the expres-
sion of DYN-1(WT) in worms treated with control RNAi (Extended Data
Fig. 6q–v,x,y). Expression of DYN-1(T67A), however, did not cause an
obvious reduction in lysosomal tubules in hpo-27-targeted RNAi worms
(Extended Data Fig. 6w,y). This result suggests that the T67A mutation
may affect the membrane fission activity of DYN-1, thereby exhibiting
a weaker tubule-suppression effect than the WT DYN-1. Expression of
DYN-1(T67A), but not the more active DYN-1(WT), exerted a differen-
tial effect on lysosomal tubules in WT and hpo-27-deficient worms.
This result implies that HPO-27 might contribute to DYN-1-dependent
lysosomal fission.
MROH1 is important for lysosomal tubule fission
In worms, MROH1::GFP formed discrete foci on lysosomes, similar
to HPO-27 (Extended Data Fig. 7l–l″). In COS7 cells, stably expressed
GFP–MROH1 was seen on vesicles and tubules labelled by the late
endosomal and lysosomal marker LAMP1 or by chased fluorescent
10 kDa dextran (Fig. 3b and Extended Data Fig. 8a–a″). Similar to
HPO-27, GFP–MROH1 formed bright foci at the neck or along the
length of lysosomal tubules labelled by LAMP1 or chased fluores-
cent dextran. Moreover, constriction and scission of the membrane
tubule occurred underneath or between the MROH1 foci (Fig. 3c,d
and Supplementary Videos 5 and 6). We generated MROH1 knock-
out COS7 cells (Extended Data Fig. 8b,c) and starved them of serum
and glutamine for 8 h to induce ALR tubules6. In MROH1 KO cells, the
number and total length of tubules were increased, and tubule fission
events were reduced (Fig. 3e–i). Therefore, MROH1 enriches at lyso-
some scission sites to promote tubule fission. We examined lysosome
acidity and proteolytic activity using Lysotracker Red, Magic Red, a
membrane-permeable cathepsin substrate33, and DQ-BSA, an endo-
cytosed probe for protease activity34. The intensity of Lysotracker
Red, Magic Red and DQ-BSA was substantially reduced in MROH1 KO
cells and MROH1-targeted siRNA knockdown cells (Extended Data
Fig. 8c–q). This result suggests that loss of MROH1 affects the acidity
and proteolytic activity of lysosomes.
Nature | www.nature.com | 3
Article
Fusion
WT cup-5(bp510) ppk-3(n2668) Fission Vesicle–tubule Tubule–tubule Retraction
d
a HPO-27::GFP (cyan) b HPO-27::GFP (cyan) c HPO-27::GFP (cyan) e f g h

Lysosomes that contain HPO-27::GFP foci (%)

100

P = 0.5174
P = 0.8454
0s 0s 0s 0s

80

NUC-1::Cherry RNST-2::Cherry SCAV-3::tagRFPt 5s 10 s 3s 3s

60

40 9s 11 s 8s 7s

Merge Merge Merge

20
16 s 14 s 17 s 10 s

p5 -5

26 -3
T

)
W

10
17 s

(b cup

68
20 s 20 s 13 s

(n ppk
WT +Psemo-1HPO-27::GFP (cyan)
WT
i hpo-27(tm5336) j Day 3 k Day 3 n o
P < 0.0001

60 25 20

tubules per unit area (μm)

Total length of lysosomal
P = 0.1244
P < 0.0001
that undergo different
dynamic changes (%)

tubules per unit area

Lysosomal tubules

20

No. of lysosomal
15

P < 0.0001
40
15
P = 0.0008

P < 0.0001
P < 0.0001

P < 0.0001
10
l Moult m Moult 10
20
5
5

0 0 0
Fission Vesicle– Tubule– Retraction WT +HPO-27 WT +HPO-27 WT +HPO-27 WT +HPO-27
tubule tubule ::GFP ::GFP ::GFP ::GFP
Fusion Day 3 Moult Day 3 Moult

Fig. 2 | HPO-27 associates with lysosomes, and loss of its function impairs j–m, Confocal fluorescence images of hypodermis in WT worms expressing
lysosomal fission. a–c, 3D reconstruction images of lysosomes in hypodermis NUC-1::Cherry without ( j,l) and with Psemo-1HPO-27::GFP (k,m) at L4 moult and
of WT (a), cup-5(bp510) (b) and ppk-3(n2668) (c) co-expressing HPO-27::GFP adult day 3. White arrowheads indicate lysosomal tubules. Blue arrowheads
driven by the semo-1 promoter and lysosomal markers NUC-1::Cherry, RNST- indicate HPO-27 foci on lysosomes. n,o, Quantification of the numbers (n; n = 11
2::Cherry or SCAV-3::tagRFPt. Boxed regions are magnified. Arrowheads worms per strain) and total length (o; left to right: n = 16, 16, 10 and 10 worms) of
indicate HPO-27–lysosome association. d, Quantification of a–c (n = 15 worms lysosomal tubules. Data show mean ± s.d. (d,i,n,o). Two-way ANOVA with
per strain). e–h, Time-lapse images of dynamic changes in fission (e), vesicle– Bonferroni post-test (i) or one-way ANOVA with Tukey’s multiple comparison
tubule (f) and tubule–tubule (g) fusion, and retraction (h) in lysosomes in WT test (d,n,o) was used to compare mutant data with WT (d,i), or WT data with
hypodermis of L4-stage larvae expressing NUC-1::Cherry. Arrows indicate and without HPO-27::GFP expression (n,o). P values are indicated. Scale bars,
tubular lysosomes. Arrowheads indicate vesicular lysosomes. i, Quantification 5 µm (a–c,e–h,j–m).
of events in e–h (n = 15 worms in each of 3 independent experiments).

interact with WT and GTP-bound active RAB-7 on lysosomes. Results

HPO-27 is recruited to lysosomes as a RAB-7 effector from in vitro pull-down assays showed that HPO-27–eGFP and MROH1–
HPO-27 lacks obvious transmembrane spans, so it may be recruited eGFP bound strongly to RAB-7(Q68L), whereas they showed weak or
to lysosomes through protein–protein and/or protein–lipid interac- no interaction with RAB-7(WT) or RAB-7(T23N) (Fig. 4k–m,o). These
tions. RAB-7 is an important small GTPase on late endosomes and lys- data indicate that HPO-27 and MROH1 act as RAB-7 effectors and are
osomes. Co-occurrence analyses showed that the majority of RAB-7 recruited to lysosomes by directly binding RAB-7.
and HPO-27 puncta were co-localized, a result consistent with their
lysosomal association (Fig. 4a,b). Overexpression of the GDP-bound
inactive RAB-7(T23N) completely disrupted the lysosomal asso- HPO-27 and MROH1 constrict and sever membrane
ciation of HPO-27, whereas overexpression of the GTP-bound active tubes
RAB-7(Q68L) substantially increased the lysosomal association of HPO- The supported membrane tube (SMrT) assay follows membrane fission
27 and the number of HPO-27 foci on lysosomes (Fig. 4c–j). MROH1 in real time36. HPO-27–eGFP and MROH1–eGFP purified from insect
also co-localized with RAB-7 on lysosomes in COS7 cells, but was cyto- cells bound strongly to phosphatidic acid (PA) on a membrane lipid
solically diffuse when RAB-7A(T22N) was overexpressed (Extended strip (Fig. 5a and Extended Data Fig. 10a). PA-containing SMrTs were
Data Fig. 8r–s‴). We analysed HPO-27–RAB-7 interactions in worms generated through the hydration of dry lipids followed by flow-induced
using a tripartite split-GFP assay35. Two short GFP peptides, GFP10 extrusion on passivated glass coverslips. Recombinant HPO-27–eGFP,
and GFP11, were fused to RAB-7 and HPO-27, respectively, whereas MROH1–eGFP or eGFP protein was passed through the membrane
the large GFP fragment, GFP1-9, was expressed as a cytosolic protein tubes in a flow chamber, and membrane scission was monitored by
(Extended Data Fig. 9a). Co-expression of GFP1-9, HPO-27::GFP11 and fluorescence microscopy. SMrTs, visualized using Texas Red DHPE, were
GFP10::RAB-7(WT) or GFP10::RAB-7(Q68L) resulted in a vesicular GFP stable throughout the entire procedure (Extended Data Fig. 10b). HPO-
pattern, which co-localized with NUC-1::Cherry. Thus, the reconstituted 27–eGFP and MROH1–eGFP efficiently coated PA-containing SMrTs, but
full-length GFP was localized to lysosomes (Extended Data Fig. 9b–c′). not phosphatidylcholine (PC)-only membrane tubes, whereas eGFP
No GFP fluorescence was seen when GFP10::RAB-7(T23N) was expressed was absent from SMrTs (Fig. 5b,c and Extended Data Fig. 10c–e). HPO-
with GFP1-9 and HPO-27::GFP11 (Extended Data Fig. 9d,d′). MROH1 27–eGFP and MROH1–eGFP formed discrete puncta along the SMrTs,
also interacted with WT and GTP-bound, but not GDP-bound, RAB-7 and constricted and severed the membrane tubes (Fig. 5b,c and Supple-
(Extended Data Fig. 9e–g′). These data suggest that HPO-27 and MROH1 mentary Videos 7 and 8). HPO-27–eGFP and MROH1–eGFP foci rapidly

4 | Nature | www.nature.com
 a 0s 1s 2s 3s 4s 5s 6s
c 0s 4s 6s

HPO-27::mKate2 SCAV-3::GFP (cyan)

GFP–MROH1 (cyan) LAMP1–Cherry

Merge

Merge
b LAMP1–Cherry GFP–MROH1 (cyan) Merge d 0s 2s 4s 6s

GFP–MROH1 (cyan) Dextran A555

Zoom Zoom Zoom

Merge

e WT
f MROH1 KO
g h i
40 100 P < 0.0001 15 P = 0.0005
P < 0.0001
Total length of lysosomal
tubules per cell (μm)

80
No. of lysosomal

fissions per cell

tubules per cell

30
No. of tubule
10
60
20
40
5
10 20

0 0 0
WT MROH1 WT MROH1 WT MROH1
KO KO KO
Fig. 3 | HPO-27 and MROH1 associate with lysosomes to promote membrane e,f, Confocal fluorescence images of WT (e) and MROH1 KO (f) COS7 cells with
fission. a,c,d, Super-resolution time-lapse images in worms co-expressing chased dextran A555 under FBS and glutamine starvation for 8 h. Arrowheads
SCAV-3::GFP and HPO-27::mKate2 (a) or in COS7 cells co-expressing GFP–MROH1 indicate lysosomal tubules labelled by chased dextran. Dashed circle indicates
and LAMP1–Cherry (c) or expressing GFP–MROH1 and carrying chased dextran the cell boundary. Boxed regions are magnified. g,h, Numbers (g) and
A555 (d). White arrows indicate lysosomal sites where HPO-27 or MROH1 is total length (h) of lysosomal tubules (n = 20 cells in each of 3 independent
concentrated and the membrane tubule is constricted. Arrowheads indicate experiments). i, Fission of lysosomal tubules was followed by time-lapse
scission. Yellow arrows indicate HPO-27 puncta are split in two when the imaging. Numbers of fission events per cell are presented (n = 10 cells in each of
underlying tubule is constricted and severed. b, Super-resolution images in 2 independent experiments). Data show mean ± s.d. (g–i). Data from MROH1 KO
COS7 cells expressing GFP–MROH1 and LAMP1–Cherry. Boxed regions are and WT were compared by Student’s two-tailed unpaired t-test. P values are
magnified. Arrows and arrowheads indicate association of MROH1 with LAMP1– indicated. Scale bars, 2 μm (a,c,d) or 5 μm (b,e,f).
Cherry-positive vesicular and tubular lysosomal structures, respectively.

formed and correlated well with regions of low tube fluorescence,
where scission occurred at multiple sites (Fig. 5b–g and Supplementary
Videos 7 and 8). On average, 2.6 and 5.2 scission events per 100 µm of
HPO-27-coated and MROH1-coated SMrTs were captured, respectively.
The tertiary structures of HPO-27 and MROH1, predicted using
RoseTTAFold, contain repetitive arrays of amphiphilic α-helix pairs
that stack into an elongated hook-like superhelical structure typical
of HEAT repeat proteins (Extended Data Fig. 10f–h). Single-molecule
imaging revealed that an average of 42 and 31 molecules were present in
each HPO-27–eGFP and MROH1–eGFP punctum on SMrTs, respectively
(Fig. 5h). Thus, HPO-27 and MROH1 may self-assemble to constrict
and sever membrane tubes. We characterized full-length HPO-27 and
MROH1 by size-exclusion chromatography coupled with multiangle
light scattering (SEC–MALS). The calculated molecular weight (MW)
of HPO-27 (about 331.5 kDa) and MORH1 (around 329.2 kDa) in solution
was much larger than the theoretical monomer MW (about 197.9 kDa
for HPO-27 and about 180 kDa for MROH1), which suggests that HPO-
27 and MROH1 tend to oligomerize (Fig. 5j,k). Negative-staining elec-
tron microscopy (EM) and 2D classification indicated that HPO-27 and
MROH1 adopt a curved linear conformation, similar to the hook-like
monomeric structure predicted by RoseTTAFold (Fig. 5l). Both proteins
self-associated into a closed ring-like structure that resembles a dimeric
conformation (Fig. 5l). We used photobleaching and single-molecule
imaging to analyse the oligomerization status of HPO-27–eGFP. Pho-
tobleaching of individual fluorophores was observed by total internal
reflection fluorescence microscopy, and the number of steps required
to bleach the fluorescent eGFPs was counted to indicate the multim-
erization status of HPO-27–eGFP. Overall, 35% of HPO-27–eGFP spots

Nature | www.nature.com | 5
Article
a Cherry::RAB-7 HPO-27::GFP (cyan) Merge b f P < 0.0001 g WT (1.8 ± 0.96, n = 60)
1.2 100 +RAB-7(Q68L) (3.6 ±

Lysosomes that contain

1.37, n = 60) P < 0.0001

contain HPO-27 puncta

P < 0.0001 50

Correlation coefficient

No. of lysosomes that

in the indicated range

80

HPO-27 foci (%)

0.9 40
60 30
0.6
40 20
P < 0.0001 0
0.3 20 10
0
c d e 0 0
0
Pearson M1 M2 None +RAB-7 +RAB-7 1–2 3–4 >4
(Q68L) (T23N) No. of HPO-27 puncta
Manders
h i j
NUC-1::Cherry NUC-1::Cherry NUC-1::Cherry

Fluorescence intensity
Fluorescence intensity

Fluorescence intensity
300 HPO-27::GFP 300 HPO-27::GFP 300 HPO-27::GFP

+Psemo-1RAB-7(Q68L) +Psemo-1RAB-7(T23N) 200 200 200

Zoom Zoom Zoom 100 100 100

0 0 0
0 2 4 6 8 0 2 4 6 8 10 0 1 2 3 4
Distance (μm) Distance (μm) Distance (μm)

(RAB-7(Q68L) vs RAB-7(WT))
IP: eGFP

Relative binding of HPO-27/

k m IP: eGFP n
IP: eGFP o ×-fold
RAB-7 RAB-7 12 P = 0.9444
MW MW RAB-7

MROH1 with RAB-7

(kDa) Trx WT Q68L T23N Trx WT Q68L T23N MW
(kDa) (kDa) Trx WT Q68L T23N
25 Flag–RAB-7 9
25 Flag–RAB-7 Flag–RAB-7
IB: Flag 25
15 Trx–Flag IB: Flag 6
IB: IB: Flag 15 Trx–Flag
250 HPO-27–
HPO-27 eGFP Trx–Flag IB: HPO-27 3 P < 0.0001
Input 15 HPO-27(ΔC)–
250 RAB-7
l MW RAB-7
IB: 250 MROH1– MBP–eGFP (WT)
eGFP eGFP 0
(kDa) Trx WT Q68L T23N HPO-27 HPO-27 MROH1
25 Flag–RAB-7 (ΔC)
15 Trx–Flag

Fig. 4 | HPO-27 is recruited to lysosomes by active RAB-7. a, Confocal
fluorescence images of WT hypodermis at adult day 1 co-expressing
Cherry::RAB-7 (ced-1 promoter) and HPO-27::GFP (semo-1 promoter). Boxed
regions are magnified. Arrowheads indicate co-localization of RAB-7 and HPO-27.
b, Quantification of a (n = 13 worms per test). HPO-27 and RAB-7 puncta were
analysed separately to produce Manders overlap coefficient M1 (RAB-7/HPO-27)
and M2 (HPO-27/RAB-7), respectively. c–e, 3D reconstruction of lysosomes in
WT adult hypodermis expressing NUC-1::Cherry and HPO-27::GFP without (c)
and with expression of RAB-7(Q68L) (d) or RAB-7(T23N) (e). Boxed regions are
magnified below (zoom). Arrowheads indicate association of HPO-27 with
lysosomes. f,g, Quantifications of c–e of the percentage (f; left to right: n = 15,
14 and 14 worms) and number (g; n = 60 lysosomes in each strain) of lysosomes.
h–j, Line scans of the lysosomal surface (dashed circles in zoom images
in c–e) in WT (h), RAB-7(Q68L) (i) and RAB-7(T23N) (j) animals. k–n, Pull-down
analyses of interactions between Flag–RAB-7 (k) and HPO-27–eGFP (l), MROH1–
eGFP (m) or HPO-27(ΔC)–eGFP (n), thioredoxin (Trx)-Flag as the control.
o, Quantification of k–n (5 independent experiments for HPO-27–eGFP and
HPO-27(ΔC)–eGFP, 3 independent experiments for MROH1–eGFP). Data show
mean ± s.d. (b,f,o). One-way ANOVA with Tukey’s multiple comparison test (f)
or Student’s two-tailed unpaired t-test (g,o) was performed to compare
datasets without and with RAB-7 expression, datasets that are linked by lines,
or MROH1 and HPO-27(ΔC) datasets with HPO-27. P values are indicated.
Supplementary Fig. 1 shows full gels. Scale bars, 5 µm (a,c–e).

were bleached in two steps, indicating dimers, whereas the remaining
spots bleached in a single step (58%) or in three steps (6%) (Extended
Data Fig. 10j–o). This result suggests that HPO-27 molecules exist in
a mixture of monomers and dimers in solution. Negative-staining
EM showed that a few HPO-27 molecules assembled into high-order
open or closed circular structures, which indicated oligomerization
(Extended Data Fig. 10i). When incubated with lipid nanotubes, HPO-27
assembled into helical structures surrounding the tube, as revealed by
negative-staining EM (Fig. 5m). To test whether HPO-27 interacts in a
head-to-tail mode, we performed tripartite split-GFP assays in worms
expressing different amino-terminal and carboxy-terminal fusions
of GFP10 and GFP11 with HPO-27 (GFP10::HPO-27, HPO-27::GFP10,
GFP11::HPO-27 and HPO-27::GFP11). In worms expressing the two
N-terminal fusions or the two C-terminal fusions, together with the
large GFP1-9 fragment, no GFP fluorescence was seen (Extended
Data Fig. 9h–i′). Expression of HPO-27::GFP10 and GFP11::HPO-27 or
HPO-27::GFP11 and GFP10::HPO-27 with GFP1-9 produced GFP puncta
that overlapped with HPO-27::mKate 2 (Extended Data Fig. 9j–k′).
Thus, HPO-27 undergoes head-to-tail homotypic assembly. MROH1
also self-interacted in a head-to-tail mode (Extended Data Fig. 9l–o′).
We attempted to disrupt the head-to-tail interaction of HPO-27 by
removing the N-terminal or C-terminal HEAT motif (HPO-27(ΔN) and
HPO-27(ΔC), respectively; Extended Data Fig. 10f). No GFP signal was
seen when these truncations were used in tripartite split-GFP assays
(Extended Data Fig. 9p–s′). When we tried to purify recombinant trun-
cated proteins, HPO-27(ΔN) severely aggregated, but a small amount of
soluble HPO-27(ΔC) was produced through fusion with MBP. Consistent
with results from the split-GFP assays, HPO-27(ΔC) lost the capacity for
self-association. The calculated MW of HPO-27(ΔC) in solution (about
278.6 kDa) was close to the theoretical monomer MW (about 267.3 kDa
with the MBP–eGFP tag) (Fig. 5j). Negative-staining EM showed that
HPO-27(ΔC) only adopted the curved linear monomeric conforma-
tion, which confirms the lack of self-association (Fig. 5l). Accordingly,
in the photobleaching and single-molecule imaging assays, 89% of
HPO-27(ΔC)–eGFP spots bleached in a single step (Extended Data
Fig. 10p–r). Thus, deletion of one C-terminal HEAT motif abolishes
the self-assembly of HPO-27. HPO-27(ΔC)–eGFP still bound PA on
membrane lipid strips but did not efficiently coat SMrTs (Fig. 5a,i).
HPO-27(ΔC)–eGFP did not form discrete GFP foci on SMrTs and did not
induce constriction or scission (Fig. 5i). These data suggest that HPO-27
and MROH1 assemble in a high-order oligomeric state on supported
membrane tubes, probably in a head-to-tail mode, and that this state
is essential for membrane constriction and scission. In worms, neither
HPO-27(ΔC)::GFP nor HPO-27(ΔN)::GFP associated with lysosomes,
and they did not rescue the lysosome collapse phenotype in hpo-27
mutants (Extended Data Figs. 2b and 7m,n). HPO-27(ΔC) still interacted
with RAB-7 in vitro, but lost the binding specificity to active RAB-7
(Fig. 4n,o). This result suggests that self-assembly is important for the
lysosomal association of HPO-27 and it may facilitate interactions of
HPO-27 with active RAB-7.
Loss of HPO-27 affects development and longevity
hpo-27 mutants are viable but exhibit developmental defects. About
13% of hpo-27(tm5336) mutants died as embryos, and 50% of hatched
embryos underwent larval arrest (Fig. 6a,b). Worms shed and

6 | Nature | www.nature.com
 a HPO-27 d f
HPO-27 (ΔC) MROH1 eGFP MBP HPO-27–
eGFP (cyan)
MROH1–
eGFP (cyan) h
TG PI Texas Red Texas Red 80

each punctum on SMrTs

DAG PI(4)P DHPE DHPE

No. of molecules in
PA PI(4,5)P2 Merge 60
Merge
PS PIP3 e g 3 1.5

MROH1–eGFP
HPO-27–eGFP
40

fluorescence
fluorescence
PE CHOL 1.0

fluorescence
1.0

fluorescence
2 1.0
PC SM

Tube
Tube
0.5 0.5 20
PG Sulfatide 1 0.5
CL Blank
0 0 0 0 0
0 3 6 9 0 5 10 15 20 HPO-27 MROH1
Length (μm) Length (μm)
b Texas Red DHPE HPO-27–eGFP (cyan) Merge j l m Nanotube only
0 HPO-27(ΔC) 278.6 kDa

31%
2 HPO-27 331.5 kDa

HPO-27–eGFP
1 700

Differential refractive

Molecular mass (kDa)

5
Time (min)

600

index (×10–5)
6
0
8 500 Zoom Zoom

69%
10 400
–1
12 300
13
–2 200

66%
c Texas Red DHPE MROH1–eGFP (cyan) Merge 30 35 40 45

MROH1–eGFP
0
Time (min)
1:00
k Nanotube + HPO-27
Time (min:s)

1:40 MROH1 329.2 kDa

34%
2:00 Molar mass
2.0 DRI 400

Molecular mass (kDa)

Differential refractive
2:40
1.5 380
index (×10–6)
3:40
4:00

HPO-27(ΔC)–eGFP
1.0 350 Zoom Zoom
i Texas Red DHPE HPO-27(ΔC)–eGFP (cyan) Merge
0

100%
0.5 330
Time (min)

4
0 300
8
15 20 25 30 35
10 Time (min)

Fig. 5 | HPO-27 and MROH1 constrict and sever SMrTs in vitro. a, HPO-27–eGFP, independent experiments. h, Photobleaching and single-molecule counting
MROH1–eGFP and HPO-27(ΔC)–MBP–eGFP bind to PA in lipid dot-blot assays. of the HPO-27–eGFP and MROH1–eGFP scaffolds on SMrTs (n = 200 foci for
Data are representative of three independent experiments. The components of each protein). j,k, Biochemical analysis of the oligomeric state of HPO-27 and
lipid are triglyceride (TG), diacylglycerol (DAG), PA, phosphatidylserine (PS), MROH1 by SEC–MALS. Calculated MWs match monomeric HPO-27(ΔC) and a
phosphatidylethanolamine (PE), PC, phosphatidylglycerol (PG), cardiolipin dimer plus monomer mixture of full-length HPO-27 and MROH1. l, Negative-
(CL), phosphatidylinositol (PI), PtdIns(4)P (PI(4)P), PtdIns(4,5)P 2 (PI(4,5)P 2), staining EM micrographs and 2D classification of full-length HPO-27 (n = 172,359),
PtdIns(3,4,5)P 3 (PIP 3), cholesterol (CHOL), sphingomyelin (SM) and sulfatide. full-length MROH1 (n = 137,570) and HPO-27(ΔC) (n = 40,987) proteins. Insets
b,c,i, Confocal fluorescence time-lapse images of HPO-27–eGFP (b), MROH1– (lower left) show magnified images of a single circular dimer of HPO-27 and
eGFP (c) and HPO-27(ΔC)–eGFP (i) on Texas Red-labelled SMrTs (89% DOPC, 10% MROH1, and a single monomer of HPO-27(ΔC). Right, main forms identified by
DOPA, 1% Texas Red-DHPE). Arrowheads indicate HPO-27 and MROH1 foci with 2D classification. m, Negative-staining EM micrographs of lipid nanotubes
constriction and scission of the underlying membrane tubes. HPO-27(ΔC)–eGFP without (top) and with (bottom) HPO-27 protein. Boxed regions below
did not form discrete foci on SMrTs. d,f, Confocal fluorescence images of are magnified images of the left and right areas, respectively. Arrowheads
HPO-27–eGFP (d) and MROH1–eGFP (f) on a Texas Red-labelled SMrT showing indicate helical HPO-27 structures assembled around the nanotube. Data are
membrane constriction (arrowheads). e,g, Line scans of HPO-27–eGFP or representative of three independent experiments. Scale bars, 5 μm (b–d,f,i)
MROH1–eGFP and tube fluorescence. Data in b–d,f,i are representative of six or 100 nm (l,m). Supplementary Fig. 1 shows full gels.

resynthesize their cuticle at four stage-to-stage moults (L1 to L4). In RNAi (Fig. 6o–r). hpo-27-targeted RNAi also increased the numbers of
this process, lysosomes are activated to facilitate cuticle replacement23. NMY-2::GFP puncta in isp-1 and eat-2 but not in daf-2 worms (Fig. 6p–r).
More than 65% of arrested hpo-27 larvae were fully enclosed within the Moreover, polyQ35::YFP aggregates accumulated at a higher level in
L1 cuticle, and the others displayed cuticle shedding defects at various worms treated with hpo-27 RNAi than in controls (Fig. 6s–u). These
larval stages (Fig. 6c,d). This result indicates that HPO-27 function is data suggest that HPO-27 is important for clearing protein aggregates.
required for cuticle replacement at moult. Accordingly, the number
and total fluorescence intensity of HPO-27::mKate2 puncta were sig-
nificantly increased at moult (Fig. 6e–h). hpo-27 mutants had a shorter Discussion
lifespan than WT worms under normal culture conditions and when Here we identified HPO-27 as a lysosome scission molecule. HPO-27
challenged with Drechmeria coniospora, a natural fungal pathogen that is recruited to lysosomes as a RAB-7 effector to mediate membrane
infects worms through the cuticle37 (Fig. 6i,j). The long-lived mutants constriction and scission. Thus, HPO-27 maintains the shape, structural
daf-2, eat-2 and isp-1 exhibit extended lifespans through reduced insulin integrity and degradation function of lysosomes, which are important
signalling, restricted caloric intake and impaired mitochondrial respi- for animal development and longevity. The human homologue MROH1
ration, respectively38–40. hpo-27-targeted RNAi reduced the lifespan of also associated with lysosomes to promote tubule fission. Our data
isp-1 and eat-2 mutants to the WT level and caused a 31% reduction in indicate that HPO-27 and MROH1 self-assemble to mediate membrane
the lifespan in daf-2 worms (Fig. 6k–m). Thus, HPO-27 is important for constriction and scission (Extended Data Fig. 10s).
maintaining normal lifespan, for survival of C. elegans under fungal
infection and for the lifespan extension in daf-2, eat-2 and isp-1 mutants. HPO-27 plays a key part in lysosomal fission
Reduced protein homeostasis, a universal hallmark of ageing, leads In hpo-27 mutants, the excessive lysosomal tubules fused to form net-
to age-dependent accumulation of misfolded and damaged proteins. works, which may lead to an increased total lysosomal volume. These
NMY-2 is an aggregate-prone protein that becomes more insoluble networks eventually collapsed in hyp7 cells, which is probably due to
with age41,42. NMY-2::GFP accumulated as puncta in WT oocytes at adult high demands for cargo degradation and catabolite recycling. hpo-27
day 5 (Fig. 6n,r). The number of NMY-2::GFP puncta was reduced in mutants accumulated membrane whorl-type lysosomes, which were
long-lived mutants but increased in worms treated with hpo-27-targeted enclosed by ribosome-decorated ER membranes and may correlate

Nature | www.nature.com | 7
Article
a b c d SCAV-3::GFP (cyan) g h
/HPO-27::mKate2
P = 0.0013 P < 0.0001 e 20 P < 0.0001 8 P < 0.0001

Total fluorescence intensity

No. of HPO-27 puncta
100 100

of HPO-27::mKate2 per
unit area (a.u., ×104)
Viable embryos (%)

L1 larvae growing
80 80 15 6

per unit area

to L4 (%)
60 60
Intermoult 10 4
40 40 f
5 2
20 20
67.3% 32.7%
0 0 0 0
WT hpo-27 WT hpo-27 Moult Inter- Moult Inter- Moult
(tm5336) (tm5336) moult moult

i j k l m
P = 0.0032
WT (20.1 ± 0.9 d) Control RNAi Control RNAi Control RNAi

P < 0.0001

P = 0.0002

P < 0.0001
hpo-27(tm5336) (15.7 ± 0.4 d) WT (82.5 ± 8.7 h) hpo-27 RNAi P < 0.0001 hpo-27 RNAi P < 0.0001 hpo-27 RNAi P < 0.0001
Control RNAi (20 ± 0.2 d) hpo-27(tm5336) P = 0.0283 daf-2;Control RNAi P < 0.0001 isp-1;Control RNAi P < 0.0001 eat-2;Control RNAi P < 0.0001
P < 0.0001

hpo-27 RNAi (15.8 ± 0.2 d) (49.7 ± 10.8 h) daf-2;hpo-27 RNAi P < 0.0001 isp-1;hpo-27 RNAi P = 0.1705 eat-2;hpo-27 RNAi P = 0.3441
100 100 100 100 100
Surviving worms (%)
Surviving worms (%)

Surviving worms (%)

Surviving worms (%)

Surviving worms (%)

80 80 80 80 80
60 60 60 60 60
40 40 40 40 40
20 20 20 20 20
0 0 0 0 0
0 10 20 30 40 0 48 96 144 0 20 40 60 80 100 0 10 20 30 40 50 0 10 20 30 40 50
Time (days) Time after infection (h) Time (days) Time (days) Time (days)

n o r 400
P < 0.0001 P = 0.0075 P = 0.0108 P = 0.3875
s u 20
P < 0.0001
P = 0.0107 P = 0.0003 P = 0.0404
No. of NMY-2::GFP puncta

No. of aggregation units

300 15

Control RNAi hpo-27 RNAi 200 Control RNAi 10

p q t
100 5

0 0
Control hpo-27 Control hpo-27 Control hpo-27 Control hpo-27
RNAi RNAi RNAi RNAi RNAi RNAi RNAi RNAi Control hpo-27
isp-1;control RNAi isp-1;hpo-27 RNAi hpo-27 RNAi RNAi RNAi
isp-1 eat-2 daf-2

Fig. 6 | Loss of HPO-27 affects animal development and longevity. isp-1(qm150) (l) and eat-2(ad116) (m) treated with control or hpo-27 RNAi
a,b, Embryonic lethality (a) and larval arrest (b) in WT animals and hpo-27(tm5336) (n = 100 worms per strain). n–u, Confocal fluorescence images (n–q,s,t) and
mutants. Data show mean ± s.d. from n = 150 embryos and n = 100 larvae in 4 quantifications (r,u) of aggregates in oocytes (n–q; adult day 5) or body wall
independent experiments. c,d, Differential interference contrast images of an muscle cells (s,t; day 2) in the indicated strains expressing NMY-2::GFP (n–q) or
arrested hpo-27(tm5336) mutant 60 h after L1 enclosed in the L1 cuticle (c; polyQ35::YFP (s,t). In n–q, arrowheads indicate NMY-2::GFP puncta, and dashed
n = 52) or exhibiting defective cuticle shedding (d; n = 52). Arrowhead indicates lines outline oocytes. Dashed circles in s and t enclose a group of Q35::YFP
unshed cuticle. e–h, Confocal fluorescence images (e,f) and quantifications aggregates, scored as one aggregation unit. For r (left to right), n = 31, 30, 30,
(g,h; mean ± s.d.) of vesicular and tubular lysosomes (white and yellow 32, 30, 34, 30 and 30 worms. For u, n = 30 worms per strain. Data show mean ± s.d.
arrowheads, respectively) in WT hypodermis at L4 intermoult (n = 12 animals) Student’s two-tailed unpaired t-test was performed to compare hpo-27 mutant
and moult (n = 13 animals). Red arrowheads indicate co-localization of HPO-27 or hpo-27 RNAi datasets with WT or control RNAi or datasets linked by lines
with lysosomes. i,j, Lifespans of WT (n = 150), hpo-27(tm5336) (n = 150), control (a,b,i,j,r,u) or moult with intermoult (g,h). In k–m, the Kaplan–Meier method
RNAi (n = 100) and hpo-27 RNAi (n = 100) worms (i), or D. coniospora-infected followed by the log-rank test was used to compare all RNAi datasets with
WT (n = 137) and hpo-27(tm5336) (n = 127) worms ( j). Three independent control RNAi, or datasets linked by lines. P values are shown. Scale bars, 5 μm
experiments were performed; mean lifespan (±s.d.) and representative lifespan (e,f,n–q,s,t) or 50 μm (c,d).
curves are shown. k–m, Lifespans of the long-lived mutants daf-2(e1370) (k),

with the collapse of lysosomal tubules. Their identity and fate await studies should address whether PA and/or these phosphoinositides
confirmation. Lysosomes form tubules to fulfil a variety of functions3. regulate HPO-27 targeting and assembly at scission sites.
In worms, catalytically active tubular lysosomes are induced at moult
to enable cuticle replacement, and tubules gradually accumulate with HPO-27 and MROH1 are self-assembling fission factors
age, concomitant with reduced lysosomal acidity, dynamics and deg- Recombinant HPO-27 and MROH1 constricted and severed SMrTs in a
radation activity23,24. At both moult and adult stages, HPO-27 associ- cell-free assay. Thus, they are sufficient to mediate membrane fission
ated with lysosomes, and its overexpression suppressed lysosomal in vitro. HPO-27 and MROH1 contain 37 HEAT motifs, which possess
tubule formation. hpo-27 mutants exhibited moulting defects and high structural flexibility and elasticity, especially when binding to
were short-lived. This finding suggests that HPO-27 acts on lysosomal partner proteins or responding to external forces21,22,44. When recruited
tubules with different properties and/or functionality to mediate fis- to lysosomes, the flexible HEAT motif arrays of HPO-27 and MROH1 may
sion, thereby playing key parts in both development and longevity. undergo structural changes, leading to the head-to-tail assembly of
Our data suggested that binding with RAB-7 recruits HPO-27 to the helical polymers encircling the tubules. The conformational changes
lysosomal surface, where other factors (membrane curvature and/or may be induced by binding to RAB-7 or other regulatory proteins and/or
charged phospholipids) may further localize HPO-27 to the scission site after membrane association through electrostatic and hydrophobic
(Extended Data Fig. 10s). As lysosomal HPO-27 foci were not disrupted in interactions. The predicted HPO-27 superhelix has multiple positively
worms defective in PPK-3 (an orthologue of PIKfyve) or CUP-5 (an ortho- charged regions on the surface, which may mediate binding with
logue of TRPML1), PtdIns(3,5)P2 and Ca2+ efflux may be dispensable in negatively charged lipids such as PA (Extended Data Fig. 10h). Our
targeting HPO-27 to scission sites. PA has been implicated in membrane data suggested that HPO-27 and MROH1 polymerization causes local
fusion and fission43, whereas PtdIns3P, PtdIns4P and PtdIns(4,5)P2 are squeezing of the tubule that may eventually lead to scission. HPO-27
involved in the exit of membrane components from lysosomes. Future and MROH1 foci resemble dynamin scaffolds assembled on SMrTs45.

8 | Nature | www.nature.com
However, dynamin utilizes GTP hydrolysis for membrane scission,
whereas HPO-27 and MROH1 severed SMrTs without the direct use of
energy. Amphipathic α-helices, which insert into one bilayer leaflet
to induce membrane bending, may also promote fission through the
hydrophobic insertion mechanism46. Moreover, HEAT repeat proteins
such as importin-β family members store energy through deformation
of the superhelix like a molecular spring; the stored energy is then
used to disassemble the nucleocytoplasmic transport complex22,44.
The high density of amphiphilic α-helices in HPO-27 and MROH1 poly-
mers may destabilize the encircled membrane tubules through shallow
hydrophobic insertion, whereas the inherent flexibility and conforma-
tional changes of the HEAT repeat superhelix enable assembly and/or
disassembly of the HPO-27 and MROH1 helical polymer for scission.
The structure of HPO-27 and MROH1 polymers, the mechanisms of
assembly, constriction and severing, and the requirement (if any) for
direct consumption of cellular energy in vivo all await future studies.
A new family of scission factors
HPO-27 and MROH1 belong to a highly conserved protein family with
orthologues in Dictyostelium, Drosophila, plants and mouse47. The
Dictyostelium and Arabidopsis orthologues associate with lysosomes
and are implicated in exocytosis of postlysosomes and regulation
of invaginated vacuolar membrane structures, respectively47,48. The
Drosophila orthologue is a RAB-7-interacting protein49. Here we found
that HPO-27 and MROH1 have identical predicated tertiary structures,
similar biochemical properties and membrane constriction and fission
activity in vitro and in vivo. They may represent a new family of scission
factors that regulate lysosomal fission. Loss of MROH1 affected lyso-
somal acidity and proteolytic activity and impaired lysosomal tubule
fission under nutrient starvation. The mechanisms underlying these
effects require in-depth study. Fission of lysosomes or lysosome-related
organelles occurs in many cellular contexts and may involve multiple
fission factors. It is important to investigate whether and how HPO-27
and MROH1 cooperate with other factors to regulate lysosome fission.
Online content
Any methods, additional references, Nature Portfolio reporting summa-
ries, source data, extended data, supplementary information, acknowl-
edgements, peer review information; details of author contributions
and competing interests; and statements of data and code availability
are available at https://doi.org/10.1038/s41586-024-07249-8.
1. Perera, R. M. & Zoncu, R. The lysosome as a regulatory hub. Annu. Rev. Cell Dev. Biol. 32,
223–253 (2016).
2. Luzio, J. P., Pryor, P. R. & Bright, N. A. Lysosomes: fusion and function. Nat. Rev. Mol. Cell
Biol. 8, 622–632 (2007).
3. Saffi, G. T. & Botelho, R. J. Lysosome fission: planning for an exit. Trends Cell Biol. 29,
635–646 (2019).
4. Carmona-Gutierrez, D., Hughes, A. L., Madeo, F. & Ruckenstuhl, C. The crucial impact of
lysosomes in aging and longevity. Ageing Res. Rev. 32, 2–12 (2016).
5. Langemeyer, L., Frohlich, F. & Ungermann, C. Rab GTPase function in endosome and
lysosome biogenesis. Trends Cell Biol. 28, 957–970 (2018).
6. Rong, Y. et al. Clathrin and phosphatidylinositol-4,5-bisphosphate regulate autophagic
lysosome reformation. Nat. Cell Biol. 14, 924–934 (2012).
7. Traub, L. M. et al. AP-2-containing clathrin coats assemble on mature lysosomes. J. Cell
Biol. 135, 1801–1814 (1996).
8. Lancaster, C. E. et al. Phagosome resolution regenerates lysosomes and maintains the
degradative capacity in phagocytes. J. Cell Biol. 220, e202005072 (2021).
9. Hirst, J. et al. Loss of AP-5 results in accumulation of aberrant endolysosomes: defining a
new type of lysosomal storage disease. Hum. Mol. Genet. 24, 4984–4996 (2015).
10. Chang, J., Lee, S. & Blackstone, C. Spastic paraplegia proteins spastizin and spatacsin
mediate autophagic lysosome reformation. J. Clin. Invest. 124, 5249–5262 (2014).
11. Boutry, M. et al. Inhibition of lysosome membrane recycling causes accumulation of
gangliosides that contribute to neurodegeneration. Cell Rep. 23, 3813–3826 (2018).
12. Sridhar, S. et al. The lipid kinase PI4KIIIβ preserves lysosomal identity. EMBO J. 32, 324–339
(2013).
13. Munson, M. J. et al. mTOR activates the VPS34–UVRAG complex to regulate autolysosomal
tubulation and cell survival. EMBO J. 34, 2272–2290 (2015).
14. Levin-Konigsberg, R. et al. Phagolysosome resolution requires contacts with the
endoplasmic reticulum and phosphatidylinositol-4-phosphate signalling. Nat. Cell Biol.
21, 1234–1247 (2019).
15. Bissig, C., Hurbain, I., Raposo, G. & van Niel, G. PIKfyve activity regulates reformation of
terminal storage lysosomes from endolysosomes. Traffic 18, 747–757 (2017).
16. Gan, Q. et al. The amino acid transporter SLC-36.1 cooperates with PtdIns3P 5-kinase to
control phagocytic lysosome reformation. J. Cell Biol. 218, 2619–2637 (2019).
17. Choy, C. H. et al. Lysosome enlargement during inhibition of the lipid kinase PIKfyve
proceeds through lysosome coalescence. J. Cell Sci. 131, jcs213587 (2018).
18. Boutry, M. et al. Arf1–PI4KIIIβ positive vesicles regulate PI(3)P signaling to facilitate
lysosomal tubule fission. J. Cell Biol. 222, e202205128 (2023).
19. Praefcke, G. J. & McMahon, H. T. The dynamin superfamily: universal membrane
tubulation and fission molecules? Nat. Rev. Mol. Cell Biol. 5, 133–147 (2004).
20. Schulze, R. J. et al. Lipid droplet breakdown requires dynamin 2 for vesiculation of
autolysosomal tubules in hepatocytes. J. Cell Biol. 203, 315–326 (2013).
21. Yoshimura, S. H. & Hirano, T. HEAT repeats—versatile arrays of amphiphilic helices
working in crowded environments? J. Cell Sci. 129, 3963–3970 (2016).
22. Kappel, C., Zachariae, U., Dolker, N. & Grubmuller, H. An unusual hydrophobic core
confers extreme flexibility to HEAT repeat proteins. Biophys. J. 99, 1596–1603 (2010).
23. Miao, R., Li, M., Zhang, Q., Yang, C. & Wang, X. An ECM-to-nucleus signaling pathway
activates lysosomes for C. elegans larval development. Dev. Cell 52, 21–37.e5 (2020).
24. Sun, Y. et al. Lysosome activity is modulated by multiple longevity pathways and is
important for lifespan extension in C. elegans. eLife 9, e55745 (2020).
25. Liu, B., Du, H., Rutkowski, R., Gartner, A. & Wang, X. LAAT-1 is the lysosomal lysine/arginine
transporter that maintains amino acid homeostasis. Science 337, 351–354 (2012).
26. Liu, Y. et al. Autophagy-dependent ribosomal RNA degradation is essential for maintaining
nucleotide homeostasis during C. elegans development. eLife 7, e36588 (2018).
27. Li, Y. et al. The lysosomal membrane protein SCAV-3 maintains lysosome integrity and
adult longevity. J. Cell Biol. 215, 167–185 (2016).
28. Tian, Y. et al. C. elegans screen identifies autophagy genes specific to multicellular
organisms. Cell 141, 1042–1055 (2010).
29. Wang, Z. et al. The Vici syndrome protein EPG5 is a Rab7 effector that determines the
fusion specificity of autophagosomes with late endosomes/lysosomes. Mol. Cell 63,
781–795 (2016).
30. Treusch, S. et al. Caenorhabditis elegans functional orthologue of human protein
h-mucolipin-1 is required for lysosome biogenesis. Proc. Natl Acad. Sci. USA 101,
4483–4488 (2004).
31. Nicot, A. S. et al. The phosphoinositide kinase PIKfyve/Fab1p regulates terminal lysosome
maturation in Caenorhabditis elegans. Mol. Biol. Cell 17, 3062–3074 (2006).
32. Marks, B. et al. GTPase activity of dynamin and resulting conformation change are
essential for endocytosis. Nature 410, 231–235 (2001).
33. Van Noorden, C. J. et al. Ala-Pro-cresyl violet, a synthetic fluorogenic substrate for the
analysis of kinetic parameters of dipeptidyl peptidase IV (CD26) in individual living rat
hepatocytes. Anal. Biochem. 252, 71–77 (1997).
34. Humphries, W. H. T. & Payne, C. K. Imaging lysosomal enzyme activity in live cells using
self-quenched substrates. Anal. Biochem. 424, 178–183 (2012).
35. Cabantous, S. et al. A new protein–protein interaction sensor based on tripartite split-GFP
association. Sci. Rep. 3, 2854 (2013).
36. Dar, S., Kamerkar, S. C. & Pucadyil, T. J. Use of the supported membrane tube assay
system for real-time analysis of membrane fission reactions. Nat. Protoc. 12, 390–400
(2017).
37. van den Boogert, P. H., Dijksterhuis, J., Velvis, H. & Veenhuis, M. Adhesive knob formation
by conidia of the nematophagous fungus Drechmeria coniospora. Antonie Van
Leeuwenhoek 61, 221–229 (1992).
38. Kenyon, C., Chang, J., Gensch, E., Rudner, A. & Tabtiang, R. A C. elegans mutant that lives
twice as long as wild type. Nature 366, 461–464 (1993).
39. Lakowski, B. & Hekimi, S. The genetics of caloric restriction in Caenorhabditis elegans.
Proc. Natl Acad. Sci. USA 95, 13091–13096 (1998).
40. Feng, J., Bussiere, F. & Hekimi, S. Mitochondrial electron transport is a key determinant of
life span in Caenorhabditis elegans. Dev. Cell 1, 633–644 (2001).
41. David, D. C. et al. Widespread protein aggregation as an inherent part of aging in C. elegans.
PLoS Biol. 8, e1000450 (2010).
42. Bohnert, K. A. & Kenyon, C. A lysosomal switch triggers proteostasis renewal in the
immortal C. elegans germ lineage. Nature 551, 629–633 (2017).
43. Zhukovsky, M. A., Filograna, A., Luini, A., Corda, D. & Valente, C. Phosphatidic acid in
membrane rearrangements. FEBS Lett. 593, 2428–2451 (2019).
44. Zachariae, U. & Grubmuller, H. Importin-β: structural and dynamic determinants of a
molecular spring. Structure 16, 906–915 (2008).
45. Dar, S., Kamerkar, S. C. & Pucadyil, T. J. A high-throughput platform for real-time analysis
of membrane fission reactions reveals dynamin function. Nat. Cell Biol. 17, 1588–1596
(2015).
46. Boucrot, E. et al. Membrane fission is promoted by insertion of amphipathic helices and is
restricted by crescent BAR domains. Cell 149, 124–136 (2012).
47. Thomason, P. A., King, J. S. & Insall, R. H. Mroh1, a lysosomal regulator localized by
WASH-generated actin. J. Cell Sci. 130, 1785–1795 (2017).
48. Hashiguchi, Y. et al. A unique HEAT repeat-containing protein SHOOT GRAVITROPISM6 is
involved in vacuolar membrane dynamics in gravity-sensing cells of Arabidopsis
inflorescence stem. Plant Cell Physiol. 55, 811–822 (2014).
49. Gillingham, A. K., Sinka, R., Torres, I. L., Lilley, K. S. & Munro, S. Toward a comprehensive
map of the effectors of Rab GTPases. Dev. Cell 31, 358–373 (2014).
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in
published maps and institutional affiliations.
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this
article under a publishing agreement with the author(s) or other rightsholder(s); author
self-archiving of the accepted manuscript version of this article is solely governed by the
terms of such publishing agreement and applicable law.
© The Author(s), under exclusive licence to Springer Nature Limited 2024
Nature | www.nature.com | 9
Article
Methods
C. elegans strains
C. elegans strains were cultured and maintained using standard
protocols. The N2 Bristol strain was used as the WT strain except
for polymorphism mapping, in which Hawaiian strain CB4856 was
used. dyn-1(ky51) was maintained at 20 °C. L4 larvae were moved
to 25 °C for 24 h before the lysosome tubulation phenotype was
examined. Transgenic animals carrying extrachromosomal arrays
(qxEx) were generated using standard microinjection methods, and
genome-integrated arrays (qxIs) were obtained by γ-ray irradiation
to achieve stable expression from arrays with low copy numbers.
Fluorescent protein knock-in at endogenous loci was accomplished
by CRISPR–Cas9-mediated genome editing. Insertion of single-copy
arrays was achieved by CRISPR–Cas9. Details of all the strains used in
this study are provided in Supplementary Data 1.
Cloning of hpo-27
The qx427, qx428, qx429, qx431, qx432, qx436, qx437 and qx458
alleles were isolated from a forward genetics screen for animals with
extensive tubular lysosomes, which were detected by labelling with
NUC-1::Cherry. These mutations failed to complement one another in
causing extensive NUC-1::Cherry-positive tubules, which suggests that
they affect the same gene. Genetic mapping and transformation rescue
experiments led to identification of the hpo-27 gene, the expression
of which fully restored the NUC-1::Cherry pattern in these mutants.
The sequence of the hpo-27 gene was determined in all hpo-27 alleles.
qx437 contains a G-to-A mutation that causes substitution of Gly316
with Glu. qx427 contains a C-to-T mutation that results in substitution
of Ala1467 with Val and a G-to-T mutation that causes a premature
stop codon at Glu1634. qx428 and qx431 carry C-to-T mutations that
cause premature stop codons at Arg551 and Gln55, respectively. qx429,
qx436 and qx458 contain G-to-A mutations that cause premature stop
codons at Trp675, Trp114 and Trp522, respectively. qx432 contains a
T-to-A mutation that causes a premature stop codon at Cys372. The
tm5336 allele contains a 348-bp deletion that removes a region from
the third intron to the fifth exon. The deletion causes a frameshift after
Asn166 and results in a premature stop codon and a truncated protein
that contains only the first 176 amino acids. The tm5336 mutant was
backcrossed with N2 worms six times before further analysis. The
X7 isoform of human MROH1 was expressed in worms and used in
sequence alignment.
Generation of knock-in worms using CRISPR–Cas9
The fluorescent tag mKate2 was knocked-in at the C terminus of
the endogenous HPO-27 coding sequence by CRISPR–Cas9. Tag
insertion by Cas9-mediated homologous recombination was per-
formed as previously described50. Two sgRNAs (sgRNA1: 5′-GCT
TGTATTCACTGGACTCGTGG-3′, sgRNA2: 5′-GTGATAAGGGAAGGGGG
AATAGG-3′) and a repair template plasmid with mKate2 containing
approximately 1-kb homology arms were used to increase the editing
efficiency. Following injection, Pmyo-2GFP was used as a selection marker
for co-conversion events in the F1 progeny. GFP-positive F1 worms were
picked and screened for tag insertion by PCR. Positive candidates were
confirmed by sequencing. All strains obtained were outcrossed with
the N2 WT strain for four times before further analysis.
Microscopy and imaging analysis
Differential interference contrast images and confocal microscopy
images were taken with an inverted laser scanning confocal microscope
(LSM 880, Carl Zeiss) with 488 nm (emission filter BP 503–530 nm) and
543 nm (emission filter BP 560–615 nm) lasers. Images were processed
and viewed using LSM Image Browser and ZEN software (v.2.3, Carl
Zeiss). Images of C. elegans and COS7 cells were taken at 20 °C and
37 °C, respectively, unless indicated otherwise in the figure legends.
Time-lapse recording and 3D reconstruction using
spinning-disk microscopy
C. elegans larvae or adults were mounted on agar pads in M9 buffer with
5 mM levamisole. Fluorescence images were captured using a ×100
objective (CFIPlan Apochromat Lambda; NA 1.45, Nikon) with immer-
sion oil (type NF) on an inverted fluorescence microscope (Eclipse
Ti-E, Nikon) with a spinning-disk confocal scanner unit (UltraView,
PerkinElmer) with 488 nm (emission filter 525 nm (W50)) and 561 nm
(dual-band emission filter 445 nm (W60)) lasers. To follow lysosome
dynamics, images were captured every 1–15 s for 1–10 min. For 3D
reconstruction, 10–15 z-series images (0.5 μm per section) were cap-
tured. The collected images were viewed and analysed using Volocity
software (v.6.3, PerkinElmer). COS7 cells were grown in confocal dishes
and imaged in DMEM without glutamine in a temperature (37 °C) and
atmosphere (5% CO2) controlled environment. To follow lysosomal
fission, images were captured every 1 s for 1 min.
Super-resolution time-lapse imaging
Grazing incidence structured illumination microscopy (GI-SIM) was
used to image the dynamics of HPO-27::mKate2 and GFP–MROH1 on
lysosomes in worms and COS7 cells, respectively. COS7 cells stably
expressing GFP–MROH1 and transiently expressing LAMP1–Cherry
or with chased fluorescent 10 kDa dextran A555 were seeded onto
Mattek glass-bottomed dishes 18–24 h before analysis. The medium
was replaced before imaging, and cells were imaged in HBSS. The raw
images of 3-phase × 3-orientation at 1.38 NA were collected using a
1.49 NA objective lens (Nikon CFI SR HP Apo total internal reflection
fluorescence (TIRF) 1003/1.49 oil) and recorded using a sCMOS camera
(Hamamatsu, Orca Flash 4.0 v.3). The GI-SIM images were typically
acquired at the imaging speed of 1 frame per 1 s (HPO-27::mKate2) or
2 s (GFP–MROH1) with 50 W cm–2 excitation intensity, and continuously
imaged for 60 time points. Every nine raw images were reconstructed
as a GI-SIM image using a SIM reconstruction algorithm51. The GI-SIM
images were analysed using ImageJ (v.2.1.4.7, NIH).
Quantification of lysosomal tubules in worms
Fluorescence images of embryos, larvae or adults expressing
LAAT-1::GFP or NUC-1::Cherry were captured using a ×63 objective on
a confocal microscope (LSM 880, Carl Zeiss). The number of lysosomal
tubules was counted within a unit area (400 μm2) in each animal. The
total length of lysosomal tubules within a unit area (50 μm2) in each
worm was measured by drawing a segmented thin line along the tubules
using the ‘line tool’ in Volocity software (v.6.3, PerkinElmer). About
10–21 animals were quantified in each strain at each stage.
Quantification of lysosome volume and lysosomal collapse
Fluorescence images of adults expressing NUC-1::Cherry (10–15 z-series
images, 0.5 μm per section) were captured by spinning-disk microscopy
(PerkinElmer). Serial optical sections were analysed, and the volume
of NUC-1::Cherry-positive lysosomes was quantified using Volocity
software (ve.6.3, PerkinElmer). A total of 14 animals were quantified
in each strain.
To quantify lysosomal collapse, fluorescence images of the hypoder-
mis in the indicated strains expressing NUC-1::Cherry, RNST-2::Cherry,
SCAV-3::tagRFPt or LMP-1::RFP were captured using an inverted confo-
cal microscope (LSM 880, Carl Zeiss). The percentage of worms exhibit-
ing reduced Cherry, tagRFPt or RFP fluorescence was quantified. About
50–100 animals were scored for each strain.
Quantification of fission and fusion events
Time-lapse images of L4 larvae expressing NUC-1::Cherry were captured
by spinning-disk microscopy (PerkinElmer). Images were captured
every 1 s for 1 min. The number of fission and fusion events was quanti-
fied. Only the initial fission or fusion event of a newly formed tubule was
scored. A total of 150 tubules from 15 animals were followed and quanti-
fied for each strain. Three independent experiments were performed.
Lysosome degradation activity assay
C. elegans lysosomes were purified by LysoIP using anti-HA magnetic
beads as previously described52. WT worms and hpo-27(tm5336) mutant
worms carrying the lysosomal membrane protein SCAV-3::RFP::3×HA
at adult day 1 were used for LysoIP experiments. Cathepsin activity
was measured in purified lysosomes using the fluorogenic cathep-
sin B and cathepsin L substrate Z-Phe-Arg-AMC (Glpbio) as previously
described52. The lysosome amount was determined by western blotting
using anti-ASP-4 antibody (Antibody Center of the Institute of Genetics
and Developmental Biology; rat; 1:1,000). To compare cathepsin activity
in WT worms and hpo-27 mutants, cathepsin activity (indicated by fluo-
rometric reading) and the lysosome amount (determined by anti-ASP-4
antibody) in WT worms were normalized to onefold and used as the
denominator to obtain the fold changes in the hpo-27 mutant strain.
In Extended Data Fig. 5a, the ratio of normalized fluorometric reading
versus lysosome amount is presented as relative cathepsin activity. Four
independent experiments were performed and quantified.
The NUC-1::Cherry cleavage assay was performed as previously
described24. The cleavage of Cherry from NUC-1 was examined by
western blotting using anti-Cherry antibodies (Proteintech, rabbit,
26765-1-AP; 1:1,000) and anti-tubulin antibodies (Sigma, rabbit, T5192;
1:5,000). The intensities of NUC-1::Cherry and Cherry bands were
quantified using ImageJ (v.2.1.4.7, NIH), and the extent of cleavage
was calculated by dividing the amount of Cherry by the total amount
of NUC-1::Cherry and Cherry. Five independent experiments were
performed and quantified for each strain.
Quantification of LGG-1 and NMY-2 puncta and polyQ35
aggregation
Fluorescence images of embryos, larvae or adults expressing GFP::LGG-1
or adults expressing NMY-2::GFP or polyQ35::YFP were captured using
a ×63 objective on a confocal microscope (LSM 880, Carl Zeiss). In
larvae and adults, the number of GFP::LGG-1 puncta was counted in a
unit area (400 μm2) of the hypodermis. In embryos, GFP::LGG-1 puncta
were counted in the whole area of the animal. A total of 14 animals were
quantified for each strain at each stage. The number of NMY-2::GFP
puncta in oocytes (the second, third and fourth oocytes counted from
the spermatheca) was counted using ZEN software (v.2.3, Carl Zeiss).
Each discrete group of polyQ35::YFP aggregates was scored as an aggre-
gation unit, and the number of polyQ35::YFP aggregation units in body
wall muscle cells was manually counted. About 30–34 animals were
quantified for each strain.
Examination of cell corpse degradation
Somatic cell corpses in the head region of living embryos at six embry-
onic stages (bean/comma, 1.5-fold, 2-fold, 2.5-fold, 3-fold, and 4-fold)
and germ cell corpses in one gonad arm of adult hermaphrodites at 48 h
after L4 were quantified. The cell corpses were identified by their raised
button-like morphology using Normarski optics. About 10–15 animals
were scored at each stage for each strain. Lysosensor staining was per-
formed as previously described53. To examine cell corpse degradation
in phagolysosomes, time-lapse images of embryos were captured by
spinning-disk microscopy (PerkinElmer). Images in 12–15 z-sections
(1.0 μm per section) were captured every 1 min for 2 h at 20 °C. Images
were viewed and analysed using Volocity software (v.6.3, PerkinElmer).
Examination of ER and mitochondrial patterns by 3D-SIM
microscopy
Fluorescence images of adult worms expressing an ER reporter or
a mitochondrial reporter were captured on a DeltaVision OMX V3
(3D-SIM) microscope (Cytiva, GE Healthcare) with a ×100/1.40 NA oil
objective (Olympus UPlanSApo), solid-state multimode lasers (488 nm)
and electron-multiplying charge-coupled device (EMCCD) cameras
(Evolve 512 × 512, Photometrics). Serial z-stack sectioning was obtained
at 125-nm intervals. To obtain optimal images, immersion oils with
refractive indices of 1.52 were used for worms on glass coverslips. The
microscope is routinely calibrated with 100 nm fluorescent spheres to
calculate both the lateral and axial limits of image resolution. SIM image
stacks were reconstructed using soft WoRx 6.1.1 (Cytiva, GE Healthcare)
with the following settings: pixel size of 39.5 nm; channel-specific opti-
cal transfer functions; Wiener filter constant of 0.0010; discard negative
intensities background; drift correction with respect to first angle;
and custom K0 guess angles for camera positions. The reconstructed
images were further processed for maximum-intensity projections
using soft WoRx 6.1.1. Pixel registration was corrected to be less than
1 pixel for all channels using 100 nm Tetraspeck beads.
Quantification of protein co-localization
Fluorescence images of the hypodermis at adult day 1 were captured by
spinning-disk microscopy (10–15 z-series images, 0.5 µm per section).
3D reconstruction images were obtained using Volocity software (v.6.3,
PerkinElmer). The percentage of lysosomes that contained HPO-27::GFP
foci was manually quantified within a unit area (400 µm2) for each
strain. To quantify co-localization of HPO-27::GFP and Cherry::RAB-7
or DYN-1::GFP with lysosomes, HPO-27::mKate2 with Mito::GFP or
GFP::TRAM-1, fluorescence images of the hypodermis at adult day 1
were captured by an inverted confocal microscope (LSM 880; Carl
Zeiss). Co-localization was measured using Pearson’s correlation coef-
ficient and Manders’ overlap using ZEN software (v.2.3, Carl Zeiss).
About 10–15 animals were quantified for each strain.
RNAi treatment
RNAi was performed by using the standard feeding method. In hpo-27
RNAi experiments, 3–5 L4 larvae (P0) were cultured on the RNAi plate,
and F1 progeny at adult stages were examined. L4440 empty vector
strain was used as a control.
Examination of embryonic and larval development and lifespan
To examine embryonic lethality, about 20 young adult worms (24 h
after L4) were placed on a plate containing nematode growth medium
(NGM) and OP50 for 1 h. The worms were then removed, and the eggs
laid on the plate were counted. Newly hatched L1 worms were trans-
ferred after 10 h and counted every 4 h until no new L1 worms hatched.
The total percentage of hatched L1 worms was quantified to determine
the level of embryonic lethality. A total of 150 embryos were quantified
for each strain.
To analyse larval arrest, 100 L1 larvae were placed on a NGM plate with
OP50 for 48 h. Worms that developed into the L4 stage were transferred
and counted every 4 h until no live larvae were present on the plate.
The total percentage of fully developed L4 worms was quantified to
determine the level of larval arrest.
Lifespan assays were performed at 20 °C as previously described54.
About 100–150 L4 worms (day 0) were placed on NGM plates with fresh
OP50 or HT115 (in RNAi experiments), 15 worms per plate. The surviving
worms were counted every 2 days and were transferred to new plates
to avoid interference from the progeny. The worms that crawled off
the plate, exploded, bagged or became contaminated were discarded.
Examination of cuticle shedding defects
Worms were cultured at 20 °C and synchronized by starvation-induced
L1-diapause. In brief, gravid adults were washed with M9 buffer and
collected in a 15-ml glass tube. Worms were then bleached and placed
in M9 buffer for 24 h to let the released embryos develop to the L1 stage.
L1 worms were transferred to a 96-well plate containing NGM and OP50,
with 20 worms per well. About 100 worms at 60 h after L1 stage were
washed by M9 buffer and mounted on 4% agarose pads with 20 mM
levamisole. Cuticle shedding defects were examined by an Axio Imager
Article
M2 microscope (Carl Zeiss). Three independent experiments were
performed in each strain.
Examination of D. coniospora infection
D. coniospora YMF1.01759 was obtained from K. Zhang (State Key Labo-
ratory for Conservation and Utilization of Bio-Resources in Yunnan,
and Key Laboratory for Microbial Resources of the Ministry of Educa-
tion, Yunnan University). The infection was performed as previously
described55. In brief, D. coniospora fungi were grown in a conical flask
containing 2% malt extract broth (Oxoid) on a rotary shaker (150 r.p.m.)
at 20 °C for 20–30 days. The conidia were collected and washed in
sterile distilled water, then the conidial suspension (108 conidia per
ml) was spread on 1.5% agar plates. After 3 days, when the conidia had
developed knobs, about 100 worms at the mid-L4 stage were added
onto the plate. After 2 h (time point 0), worms were transferred to NGM
plates spotted with OP50. The surviving worms were counted every 24 h
and were transferred to new NGM plates to avoid interference from their
progeny. The worms that crawled off the plate were discarded. Three
independent experiments were performed in each strain.
TEM
TEM analysis was performed as previously described56. Worms at day 1–3
of adulthood were rapidly frozen, substituted and electron-stained.
The embedded samples were cut into 70-nm thick sections with a
microtome (EM UC7, Leica Biosystems). HT7800 (Hitachi) and JEM-
1400 ( Jeol) were used for image acquisition at 80 kV. For quantitative
analyses of lysosomes, three to seven worms were analysed for each
strain at each stage. Three to five 70-nm sections (non-consecutive
sections, spaced at 3,000 nm) were analysed in each animal. In each sec-
tion, the total surface area of the epidermal syncytium and lysosomes
were measured using ImageJ software (v.2.1.4.7, NIH).
TEM with DAB labelling was performed as previously described57. In
brief, adult worms were fixed, and after rehydration, samples were cut
off with a blade and stained by DAB. After DAB staining, samples were
fixed, dehydrated and gradually infiltrated. The embedded samples
were cut into 70-nm sections with a microtome EM UC7 (Leica) and
electron-stained with uranyl acetate and lead citrate. Sections were
observed with a HT7800 (Hitachi) operating at 80 kV.
Cell culture
COS7 cells (CRL-1651, American Type Culture Collection) were grown
in DMEM (10569010, Gibco) supplemented with 10% FBS (16000044,
Gibco) and penicillin–streptomycin (100 mg ml–1, 15140122, Gibco)
at 37 °C with 5% CO2. This cell line is not listed as misidentified or
cross-contaminated in the International Cell Line Authentication Com-
mittee database, and is free of mycoplasma contamination. For serum
and glutamine starvation, cells were washed three times with PBS and
incubated with DMEM without glutamine (11960044, Gibco) for 8 h
before imaging. Whenever thawed, cells were passaged at least three
times before being used in experiments.
Generation of COS7 cells with stable expression of GFP–MROH1
The COS7 cell line with stable expression of GFP–MROH1 was gen-
erated by lentiviral transduction. GFP–MROH1 was subcloned into
pCDH-EF1-MCS-Puro. Lentivirus particles were packaged using a
second-generation lentiviral packaging system (psPAX2 and PMD2.G).
Cells were then transduced with a low virus titre, and stable cells were
selected by puromycin (2 μg ml–1). The selected cells were further sorted
by FACS into 96-well plates (1 cell per well). Clones that grow well and
had clear GFP–MROH1 expression were used for experiments.
Generation of the CRISPR-mediated MROH1 knockout COS7 cell
line
sgRNAs targeting the green monkey MROH1 genomic sequence
were designed using CRISPick (https://portals.broadinstitute.org/
gppx/crispick/public) and cloned into pSpCas9(BB)-2A-GFP (PX458,
Addgene, 48138). sgRNA-encoding plasmids were transfected into
0.5 × 106 COS7 cells using Lipofectamine 2000 (12566014, Life Tech-
nologies). Following 3 days of transfection, GFP-positive cells were
sorted by FACS and seeded into 96-well plates by serial dilution. Clonal
cells were then expanded for genome extraction and DNA sequencing.
Transfection and RNAi
Lipofectamine 3000 (100022052, Invitrogen) was used for transient
transfection of plasmids. For siRNA interference, cells were transfected
using Lipofectamine RNAi MAX (13778150, Life Technologies) with a
final siRNA concentration of 100 nM. Cells were cultured for 3 days
before analysis. MROH1 and control siRNA oligonucleotides were syn-
thesized by GenePharma.
Labelling of lysosomes by fluorescent dextran
The dextran chase experiment was performed as previously described18.
Cells were incubated with 0.1 mg ml–1 of fluorescent 10 kDa dextran A555
(D34679, Thermo Fisher Scientific) for 3 h at 37 °C. Cells were washed
twice and chased overnight in DMEM containing 10% FBS to allow the
dextran to reach lysosomes before analysis. In serum and glutamine
starvation experiments, cells containing overnight-chased dextran
were washed three times with PBS and cultured in DMEM without glu-
tamine (11960044, Gibco) for 8 h before analysis.
Analysis of lysosomal tubules and tubule fission in COS7 cells
COS7 cells were cultured in DMEM without glutamine for 8 h. Fluo-
rescence images of lysosomes labelled by chased dextran A555 were
captured using a ×100 objective on a spinning-disk confocal micro-
scope (PerkinElmer) at 37 °C with 5% CO2. The number of lysosomal
tubules per cell was manually counted, and the total length of tubules
was measured by drawing a segmented thin line along the tubules using
the ‘line tool’ in Volocity software (v.6.3, PerkinElmer). At least 20 cells
were quantified each time in each cell line, and 3 independent experi-
ments were performed.
To follow fission of lysosomal tubules, time-lapse images of COS7
cells with chased dextran A555 were captured by spinning-disk confocal
microscopy (PerkinElmer) at 37 °C and with 5% CO2. The frames were
taken every 1 s for 1 min. The number of fission events occurring at the
neck or in the middle of lysosomal tubules was counted in each cell. A
total of 20 cells from 2 independent experiments were quantified in
each cell line.
Lysosome assays
The following staining assays were performed to examine the acid-
ity and proteolytic activity of lysosomes. COS7 cells seeded on a
glass-bottom dish were incubated with LysoTracker Red (L7528, Inv-
itrogen, 100 nM) or Magic Red L (ICT-941, Immunochemistry, 1:250) at
37 °C for 30 min, or DQ BSA (D12051, Invitrogen, 4 μg ml–1) at 37 °C for
8 h. Cells were washed with medium and live-imaged immediately at
37 °C with 5% CO2 using an inverted laser scanning confocal microscope
(LSM 880, Carl Zeiss). Four independent experiments were performed
in WT and MROH1 KO cells and three independent siRNA experiments
were performed.
Quantitative real-time PCR
Total RNA was extracted from worms or COS7 cells using TRIzol
(15596018CN, Invitrogen) and reverse transcribed using a PrimeScript
RT Reagent kit (RR037A, Takara) according to the manufacturer’s
protocol. The reverse transcription products (cDNA) were diluted to
100 ng μl–1, and 2.5 μl of the diluted cDNA was used as the template
for quantitative PCR in a 20-μl reaction mixture. For quantitative PCR
with reverse transcription, FastStart Universal SYBR Green Master
(4913914001, Roche) was used on an Applied Biosystems QuantStudio
3 Flex real-time PCR system (Life Technologies). The GAPDH gene and
the act-1 gene were used as the internal reference for COS7 cells and
worms, respectively. Three independent experiments were performed
with three replications each time.
Protein expression and purification
The cDNA of C. elegans RAB-7 was inserted into a modified pET28a vec-
tor with an N-terminal Flag-tag and His6-tag. For HPO-27 and MROH1,
the cDNA sequences encoding full-length protein and HPO-27(ΔC)
(residues 1–1725), or MROH1 (residues 94–1725) were each cloned into
a modified pFastBac1 (Invitrogen) vector. The resulting constructs
contain a C-terminal eGFP-His6-strep tandem tag. For HPO-27(ΔC), an
extra MBP tag was added to the C terminus of the protein to improve
solubility. The standard PCR-based mutagenesis method was used to
generate the mutations in RAB-7. All the constructs were confirmed by
DNA sequencing. For RAB-7, the recombinant proteins were expressed
in Escherichia coli BL21(DE3) host cells at 37 °C. The WT and mutant
proteins were purified by Ni2+-Sepharose 6 Fast Flow (GE Healthcare)
affinity chromatography. The purified proteins were dialysed into a
buffer containing 25 mM HEPES, pH 7.5, 150 mM NaCl, 5 mM MgCl2.
For the Q68L and T23N mutant proteins, 0.1 mM GTPγS or 0.1 mM GDP
was added in the purification buffer, respectively. The proteins were
concentrated to 2 mg ml–1 and stored at −80 °C. For HPO-27, HPO-27(ΔC)
and MROH1, the recombinant proteins were expressed in insect sf9
(11496015, Gibco) cells. The cells were collected and suspended in a
buffer containing 50 mM Tris-HCl, pH 8.0, 500 mM NaCl, 10% glycerol
and 5 mM imidazole. The proteins were purified by Ni2+-Sepharose 6
Fast Flow (GE Healthcare) affinity chromatography followed by
Strep-Tactin Sepharose column chromatography (IBA Lifesciences).
The eluted proteins were further purified by size-exclusion chromatog-
raphy (Superdex-200 10/300, GE Healthcare) with a buffer containing
50 mM Tris-HCl, pH 8.0, and 150 mM NaCl. For the pull-down assay,
the proteins were dialysed into a buffer containing 25 mM HEPES,
pH 7.5, 150 mM NaCl.
SEC–MALS
Protein samples (about 1 mg ml–1 in 50 mM Tris-HCl, pH 8.0, and
150 mM NaCl) were analysed with static light scattering by inject-
ing them into an Agilent FPLC system with a Superdex-200 increase
10/300 column (GE Healthcare). The chromatography system was
coupled with an 18-angle light-scattering detector (Dawn Heleos
II, Wyatt Technology) and a differential refractive index detector
(Optilab rEx, Wyatt Technology). Masses (MWs) were calculated using
ASTRA (Wyatt Technology). BSA (Sigma) was used as the calibration
standard.
GFP-nanobody pull-down assay
A total of 1.5 μg Flag-tagged RAB-7 (WT, Q68L and T23N) or Trx-Flag
(the control) was mixed with 5 μg eGFP-tagged HPO-27, MROH1 or HPO-
27(ΔC) in reaction buffer (25 mM HEPES, pH 7.5, 150 mM NaCl and 5 mM
MgCl2) containing either 0.1 mM GTPγS or 0.1 mM GDP. Reactions were
incubated on ice for 1 h and then 5 μM chemical cross-linker DSP (22585,
ThermoFisher) was added in the reactions and incubated at room tem-
perature for 30 min. The reactions were quenched with 50 mM Tris-HCl,
pH 8.0. Next, 5 μl GFP-Trap agarose (gta-10, Chromotek) was added
in each reaction and incubated at 4 °C for 30 min. After incubation,
the beads were washed 4 times in a buffer containing 25 mM HEPES,
pH 7.5, 500 mM NaCl, 5 mM MgCl2, 0.25% Triton X-100 with 0.1 mM
GTPγS or 0.1 mM GDP. The beads were finally boiled in 20 μl loading
buffer and the resulting samples were separated by SDS–PAGE and
analysed by western blotting using anti-Flag (Sigma, mouse, F3165;
1:1,000), anti-HPO-27 (Antibody Center of National Institute of Bio-
logical Sciences, mouse; 1:1,000) or anti-GFP (Abcam, rabbit, ab290;
1:1,000) antibodies. The intensity of the protein band was quantified
using ImageJ (v.2.1.4.7, NIH). Three to five independent experiments
were performed.
Protein prediction, negative-staining EM, data collection and
2D classification
The structures of full-length HPO-27 and human MROH1 were predicted
using RoseTTAFold (https://robetta.bakerlab.org)58. Comparisons
and illustrations were prepared using the program PyMOL (https://
pymol.org).
Protein samples of HPO-27–eGFP, MROH1–eGFP or HPO-27(ΔC)–
MBP–eGFP were diluted to 0.05 mg ml–1 with a buffer containing 50 mM
Tris-HCl, pH 8.0, and 150 mM NaCl. To examine HPO-27 in the presence
of lipid nanotubes59, Rhod-labelled liposomes (40% glucosyl(β) cera-
mide (SPHZ-082, Creative Enzymes), 59% DOPA (840875P, Avanti Polar
Lipids) and 1% Rhod-DOPE (810150P, Avanti Polar Lipids)) were dried to
a film, then resuspended in 50 mM Tris-HCl, pH 8.0, and 150 mM NaCl
followed by incubation with a final concentration of 0.015 mg ml–1 HPO-
27–eGFP. Each sample was adsorbed onto a freshly glow-discharged
carbon film and mounted on a 200-mesh EM grid. After 1 min, the
samples were washed once in a 2% filtered solution of uranyl acetate
and then stained for 1 min. Electron micrographs were manually col-
lected in a Tecnai Spirit electron microscope (FEI) operated at 100 kV,
with a magnification of ×98,000, a defocus range of −0.5 to −2.5 μm
and a pixel size of 5.78 Å per pixel. All the 2D classification processing
was done using RELION 3.1.3 (Sjors Scheres Laboratory, MRC Labora-
tory of Molecular Biology). About 500–1,500 particles were manually
picked and extracted from the micrographs for 2D classification. The 2D
classification models were then chosen as autopicked templates. The
particles were next autopicked and extracted from the micrographs.
A total of about 290,000, 340,000 and about 46,000 particles were
extracted for HPO-27, MROH1 and HPO-27(ΔC), respectively. Twenty
classes were obtained for HPO-27, MROH1 and HPO-27(ΔC) after three
to six rounds of 2D classification.
Lipid dot-blot assay
Membrane lipid strips (P-6002) were obtained from Echelon. Protein
(10 nM) was incubated with pre-blocked lipid strips at 4 °C for 3 h.
Membrane lipid strips were washed and blotted with anti-GFP (Abcam,
rabbit, ab290; 1:1,000), anti-MBP (NEB, mouse, E8032; 1:5,000) or
anti-HPO-27 antibodies (Antibody Center of National Institute of Bio-
logical Sciences, mouse; 1:1,000). Three independent experiments were
performed, and a representative result is shown in Fig. 5a.
SMrT-based membrane fission assay
The SMrT-based membrane fission assays were performed as previously
described with modifications35,44. In brief, lipids (DOPC (850375P, Avanti
Polar Lipids), DOPA (840875P, Avanti Polar Lipids) and Texas Red DHPE
(T1395MP, Invitrogen) at a mole percentage of 89:10:1) were diluted in
chloroform to a final total lipid concentration of 1 mM. A small aliquot
(about 1 μl) was spread on a freshly cleaned PEG8000-coated glass
coverslip and kept under high vacuum for 30 min to remove traces
of chloroform. A flow cell (FCS2 system, Bioptechs) was assembled
by placing a piece of 0.1-mm silicone spacer between the PEGylated
coverslip and an ITO-coated glass slide. The flow cell was filled with
50 mM Tris-HCl, pH 8.0, and 150 mM NaCl and left undisturbed for
10 min at room temperature. SMrTs were generated by extrusion of
the large vesicles formed during hydration to narrow membrane tubes
by flowing a large volume of buffer at high flow rates. Then, 200 μl of
0.2 µM HPO-27–eGFP, HPO-27(ΔC)–eGFP, MROH1–eGFP or eGFP was
flowed into the chamber at a low flow rate. Fluorescence images were
subsequently taken by an inverted confocal microscope (LSM 880,
Carl Zeiss).
Photobleaching and single-molecule imaging assay
To determine the molecular stoichiometry of HPO-27 within protein
complexes, the photobleaching and single-molecule imaging assay was
performed as previously described60. Applying a defined continuous
Article
laser power to the protein solution of eGFP, HPO-27–eGFP, HPO-27(ΔC)–
eGFP or the SMrTs with assembled HPO-27–eGFP or MROH1–eGFP scaf-
folds causes stepwise loss of the GFP fluorescence by photobleaching,
which was captured by a TIRF microscope. For each chosen fluorescent
spot, the number of bleaching steps was counted to determine the
number of eGFP molecules it contains. To do this, 1 nM HPO-27–eGFP,
HPO-27(ΔC)–MBP–eGFP or eGFP proteins were adsorbed onto a freshly
glow-discharged and cleaned coverslip followed by imaging with a
TIRF microscope (Olympus). Proteins in solution or on the SMrTs were
excited with a 488-nm laser, and 160 W cm–2 excitation intensity was
applied to the rear pupil of an oil objective (Olympus ×150, N.A. = 1.45).
The frames were taken every 0.05 s for 500–1,000 frames. To reduce
background noise and maximize emission, photons were detected
using a back-illuminated EMCCD camera (Andor iXon DV-897 BV)
and high-quality filters (Chroma, 49002 ET-GFP). In addition, a laser
blocker absorbed the reflected laser beam at the rear pupil of the
objective. The gain of the EMCCD camera was set at 300 throughout
all single-molecule imaging experiments. The centre of mass of each
fluorescent punctum was determined by a previously developed algo-
rithm written in Matlab (v.R2017b, MathWorks)61. To quantify HPO-27
and MROH1 molecules assembled on SMrTs, we first determined the
photons required to bleach a single eGFP molecule. Then, we collected
the initial photons of each HPO-27–eGFP or MROH1–eGFP punctum
formed on SMrTs. The number of HPO-27 or MROH1 molecules in each
punctum was calculated by dividing the total initial photons of each
HPO-27–eGFP or MROH1–eGFP punctum by the photons of a single
eGFP molecule.
Expression vectors
Expression vectors and oligonucleotides used in the study are provided
in Supplementary Data 2 and 3.
Statistical analysis
The standard deviation or standard error of the mean was used as
the y axis error bar for bar charts plotted from the mean value of the
data. Data derived from different genetic backgrounds or different
conditions were compared using Student’s two-tailed unpaired t-test,
one-way ANOVA followed by Tukey’s multiple comparison test, two-way
ANOVA followed by Bonferroni post-test or the Kaplan–Meier method
followed by log-rank test as indicated in the figure legends. Data were
considered statistically different at P < 0.05. P values are shown in
graphs. GraphPad Prism (v.8.2.1, Dotmatics) was used for statistical
analyses.
Reporting summary
Further information on research design is available in the Nature Port-
folio Reporting Summary linked to this article.
Data availability
All data are available in the main text and supplementary materials.
Structural prediction was performed using RoseTTAFold (https://
robetta.bakerlab.org) and illustrations were prepared using the pro-
gram PyMOL (https://pymol.org). Images were processed using ZEN
Imaging software (v.2.3, Carl Zeiss), ImageJ (v.2.1.4.7, NIH), Volocity
software (v.6.3, PerkinElmer), RELION (v.3.1.3, Sjors Scheres Laboratory,
MRC Laboratory of Molecular Biology), WoRx (v.6.1.1, GE Healthcare)
and Matlab (v.R2017b, MathWorks). Statistical analyses were performed
using GraphPad Prism (v.8.2.1, Dotmatics). Full versions of all gels and
blots are provided in Supplementary Fig. 1. Source data are provided
with this paper.
Code availability
No custom code was generated or applied in this study.
50. Paix, A. et al. Scalable and versatile genome editing using linear DNAs with microhomology
to Cas9 Sites in Caenorhabditis elegans. Genetics 198, 1347–1356 (2014).
51. Guo, Y. et al. Visualizing intracellular organelle and cytoskeletal interactions at nanoscale
resolution on millisecond timescales. Cell 175, 1430–1442.e17 (2018).
52. Zhang, Q., Li, Y., Jian, Y., Li, M. & Wang, X. Lysosomal chloride transporter CLH-6 protects
lysosome membrane integrity via cathepsin activation. J. Cell Biol. 222, e202210063
(2023).
53. Guo, P., Hu, T., Zhang, J., Jiang, S. & Wang, X. Sequential action of Caenorhabditis elegans
Rab GTPases regulates phagolysosome formation during apoptotic cell degradation.
Proc. Natl Acad. Sci. USA 107, 18016–18021 (2010).
54. Hansen, M., Hsu, A. L., Dillin, A. & Kenyon, C. New genes tied to endocrine, metabolic,
and dietary regulation of lifespan from a Caenorhabditis elegans genomic RNAi screen.
PLoS Genet. 1, 119–128 (2005).
55. Jansson, H. B. Adhesion of conidia of Drechmeria coniospora to Caenorhabditis elegans
wild type and mutants. J. Nematol. 26, 430–435 (1994).
56. Li, Y., Wang, X., Li, M., Yang, C. & Wang, X. M05B5.4 (lysosomal phospholipase A2)
promotes disintegration of autophagic vesicles to maintain C. elegans development.
Autophagy 18, 595–607 (2022).
57. Tsang, T. K. et al. High-quality ultrastructural preservation using cryofixation for 3D electron
microscopy of genetically labeled tissues. eLife 7, e35524 (2018).
58. Baek, M. et al. Accurate prediction of protein structures and interactions using a three-track
neural network. Science 373, 871–876 (2021).
59. Kulkarni, V. S., Anderson, W. H. & Brown, R. E. Bilayer nanotubes and helical ribbons
formed by hydrated galactosylceramides: acyl chain and headgroup effects. Biophys. J.
69, 1976–1986 (1995).
60. Ji, W. et al. Functional stoichiometry of the unitary calcium-release-activated calcium
channel. Proc. Natl Acad. Sci. USA 105, 13668–13673 (2008).
61. Ulbrich, M. H. & Isacoff, E. Y. Subunit counting in membrane-bound proteins. Nat. Methods
4, 319–321 (2007).
Acknowledgements We thank S. Mitani (Tokyo Women’s Medical University) and staff at the
Caenorhabditis Genetics Center for C. elegans strains used in this study; K. Zhang (Yunnan
University) for Drechmeria coniospora fungi; Y. Li, X. Guo, M. Wang, X. Ren, X. Su, H. Wang
(Institute of Biophysics, Chinese Academy of Sciences), X. Jia and S. Li (Center for Biological
Imaging, Institute of Biophysics, Chinese Academy of Sciences), Y. Guo and X. Wu (Institutional
Center for Shared Technologies and Facilities, Kunming Institute of Zoology, Chinese Academy
of Sciences), Q. Yin and X. Wang (School of Life Sciences, Yunnan University) for technical
assistance; and I. Hanson for editing services. This study was equally supported by grants to
X.W. from the National Natural Science Foundation of China and National Key R&D Program of
China (92254303, 2021YFA1300303, 32293202 and 32130028); and grants to W.F. from the
National Natural Science Foundation of China (32071191) and Strategic Priority Research
Program of CAS (XDB37020302).
Author contributions X.W. supervised the study. L.L. performed most of the experiments.
X.L., Y.W. and S.H. performed genetic and cell biology experiments. W.F. supervised, and S.Y.
and W.W. performed, protein characterization and negative-staining EM experiments. C.Y.
supervised, and M.L. performed, TEM analysis. D.L. supervised, and A.J. and X.M. performed,
GI-SIM experiments. Junjie Hu analysed structure prediction data and supervised in vitro
reconstitution experiments with X.W. J.Z. and Junyan Hu generated COS7 KO and stable
expression cell lines and contributed to the COS7 cell experiments. Q. Zhang, Q. Zhao and Y.L.
contributed to the experiments and materials. L.L., X.L., W.F. and X.W. prepared the manuscript
with discussion among all authors.
Competing interests The authors declare no competing interests.
Additional information
Supplementary information The online version contains supplementary material available at
https://doi.org/10.1038/s41586-024-07249-8.
Correspondence and requests for materials should be addressed to Wei Feng or Xiaochen
Wang.
Peer review information Nature thanks Roberto Botelho, Thomas Pucadyil, Luca Scorrano and
Nektarios Tavernarakis for their contribution to the peer review of this work.
Reprints and permissions information is available at http://www.nature.com/reprints.
Extended Data Fig. 1 | hpo-27 mutants contain extensive lysosomal tubules.
a-r, Confocal fluorescence images of the hypodermis in day 2 adult worms
carrying different hpo-27 mutant alleles (a-p), or treated with hpo-27 RNAi (q, r).
The worms were expressing NUC-1::CHERRY and AJM-1::GFP to label lysosomes
and lateral seam cells, respectively. hpo-27 mutants contain extensive lysosomal
tubules and greatly diminished NUC-1::CHERRY fluorescence in hyp7 cells. The
percentage of worms with the representative lysosomal pattern is shown at the
lower left corner in each panel. s, qPCR analysis of hpo-27 gene expression in
control RNAi and hpo-27 RNAi worms. Three independent experiments were
performed, and data are shown as mean ± SD. Student’s two-tailed unpaired
t-test was performed to compare hpo-27 RNAi with control RNAi. P values are
indicated. t-v”, Confocal fluorescence images of the hypodermis in wild type
(WT) (t-t”) and hpo-27(tm5336) mutant adults (u-v”) carrying NUC-1::CHERRY
and Psemo-1GFP (n = 15 worms in each strain). Lysosomes labeled by NUC-1::CHERRY
are seen in both hyp7 cells positive for GFP and lateral seam cells that lack GFP
expression (arrows). In hpo-27 mutants, extensive lysosomal tubules are
present in hyp7 and seam cells, and NUC-1::CHERRY fluorescence is diminished
in hyp7 but not seam cells. w, Cloning of hpo-27. The hpo-27 gene structure is
shown with filled boxes representing exons, thin lines indicating introns, and
gray bars designating 3’ and 5’ UTRs. The arrow delineates the direction of
transcription. x, Schematic diagram showing the organization of the HPO-27
protein. The mutation sites identified in all the hpo-27 alleles and the region
deleted in the tm5336 allele are indicated. Gray boxes indicate the HEAT motifs.
Scale bars: 5 µm.
Article
Extended Data Fig. 2 | HPO-27 is homologous to human MROH1. a, Sequence
alignment of C. elegans (c.e) HPO-27 and human (h.s) MROH1. Identical residues
are shaded in black. The red box indicates the disordered region and blue boxes
designate the HEAT motifs. b, The percentage of worms with diminished
NUC-1::CHERRY fluorescence was quantified in the indicated strains at adult
day 3 (n = 50 worms in each of 3 biological repeats for each strain). Data are shown
as mean ± SD. One-way ANOVA with Tukey’s multiple comparison test was
performed to compare mutant datasets with WT or datasets that are linked by
lines. P values are indicated.
Extended Data Fig. 3 | Loss of HPO-27 affects lysosome morphology in
intestine and body wall muscle cells. a-d, Confocal fluorescence images of
the intestine and body wall muscle cells in WT and hpo-27(tm5336) adult worms
expressing LAAT-1::GFP. Arrowheads indicate tubular lysosomes. Quantifications
are shown in e, f, h (n = 14 worms per strain) and g (n = 15 worms per strain). Data
are shown as mean ± SD. Two-way ANOVA with Bonferroni post-test (e, f, h) or
one-way ANOVA with Tukey’s multiple comparison test (g) was performed to
compare mutant datasets with WT. P values are indicated. i-q, Confocal
fluorescence images of the hypodermis in WT and hpo-27(tm5336) worms at
adult day 2 expressing AJM-1::GFP (which stains the seam cell boundary) and
RNST-2::CHERRY (i-k) or SCAV-3::tagRFPt (l-n), or carrying Pcol-12LMP-1::RFP
which is expressed in hyp7 but not seam cells (o-q). The percentage of worms
with the representative lysosomal pattern is shown at the lower left corner in
each panel. In (q), the DIC image is merged to show a hyp7 cell with invisible
LMP-1::RFP. r-v’, Confocal fluorescence time-lapse images of the hypodermis in
adult hpo-27(tm5336) worms co-expressing RNST-2::CHERRY and SCAV-3::GFP.
The lysosomal tubular networks are clearly observed at “0 min”, but they
gradually collapse. White arrowheads indicate the lysosomal tubules that
gradually disappear. The lysosomal membrane protein SCAV-3::GFP labels
tubular structures, which collapse to form small vesicles (yellow arrowhead)
and eventually became diffuse. Scale bars: 5 µm.
Article

Extended Data Fig. 4 | See next page for caption.

Extended Data Fig. 4 | Loss of HPO-27 affects lysosomal integrity and
causes abnormalities of the ER and mitochondria. a-i, Transmission electron
micrographs of the hypodermis in WT and hpo-27 RNAi worms expressing
SCAV-3::APEX. Both “membrane whorl”-containing compartments and tubules
are stained by diaminobenzidine (DAB) (arrowheads). j-n, Transmission electron
micrographs of the hypodermis in hpo-27(tm5336) at adult day 3. Few lysosomes
(green arrowhead) are observed, and they are surrounded by ribosome-
containing ER membranes (yellow arrowheads in k-n). Red arrowheads indicate
membranes of autophagosomes (AP) enclosing a mitochondrion or an ER
microsome (yellow asterisks). The pink arrowhead indicates a piece of broken
ER. “M” indicates a fragmented and dilated mitochondrion. Quantification of
the area of hypodermal lysosomes is shown in o (left to right: n = 4, 3, 4, 6 worms).
Data are shown as mean ± SEM. Student’s two-tailed unpaired t-test was
performed to compare mutant datasets with WT. P values are indicated.
p-aa, Super-resolution images of the hypodermis in the indicated strains
expressing the ER marker GFP::TRAM-1 (p-u) or the mitochondrial marker
Mito::GFP (v-aa) at adult day 1 and day 3. Arrowheads indicate abnormal
ring-like ER structures. The percentage of worms with the representative ER
or mitochondrial pattern is shown at the lower left corner in each panel.
ab-ae, Transmission electron micrographs of the hypodermis in WT and
hpo-27(tm5336) at adult day 1 and day 3. Pink arrowheads indicate normal ER
and yellow arrowheads indicate microsomes probably generated by structural
damage of the ER. “M” indicates a mitochondrion, and yellow asterisks designate
abnormal mitochondria. af, ag, Confocal fluorescence images of the hypodermis
in WT expressing HPO-27::mKate2 and Mito::GFP or GFP::TRAM-1. The boxed
region is magnified in the zoom images. Magenta and white arrowheads
indicate HPO-27::mKate2 puncta and mitochondrial or ER structures,
respectively. Quantification is shown in ah (n = 10 worms per strain). Data are
shown as mean ± SD. Scale bars represent 0.5 µm in (a-n, ab-ae) and 5 µm in
(p-aa, af, ag).
Article
Extended Data Fig. 5 | Loss of hpo-27 affects lysosomal degradation activity.
a, Relative activity of cathepsin B and L in lysosomes purified from WT and hpo-
27(tm5336) mutants. The amount of lysosomes in each strain was determined
by western blot with an anti-ASP-4 antibody. b, Western blot analysis of NUC-
1::CHERRY cleavage in WT and hpo-27(tm5336) at day 1. Quantification is shown
in c. Four (a) and five (c) independent experiments. d-r, Confocal fluorescence
images of WT, epg-5(tm3425), and hpo-27(tm5336) animals expressing GFP::LGG-1.
Comma and 4-fold are embryonic stages; day 1 and day 3 are adult stages. LGG-1
puncta are indicated by white arrowheads. Quantifications are shown in s
(n = 14 worms per strain). t-w, DIC images of the embryo (t, v) and germline
(u, w) in WT and hpo-27(tm5336) mutants. Cell corpses are indicated by
arrowheads. Quantifications are shown in x (n = 10 worms per strain), y (n = 15
worms per strain). Comma, 1.5 F, 2 F, 2.5 F, 3 F and 4 F (4-fold) are different
embryonic stages. L4 is larval stage 4. In (a, c, s, x, y), data are shown as mean ± SD.
Two-way ANOVA with Bonferroni post-test (s, x) or Student’s two-tailed unpaired
t-test (a, b, y) was performed to compare mutant datasets with WT. P values are
indicated. Scale bars: 5 µm. For gel source data, see Supplementary Fig. 1.
Extended Data Fig. 6 | See next page for caption.
Article
Extended Data Fig. 6 | dyn-1 mutants exhibit a weak lysosomal tubule
phenotype. a-b’, DIC and confocal fluorescence images of the gonads in WT (a,
a’) and hpo-27(tm5336) (b, b’) stained by Lysosensor green. White and yellow
arrowheads indicate cell corpses positive and negative for Lysosensor green,
respectively. Quantification is shown in c (n = 15 worms per strain). d, Confocal
time-lapse images of WT and hpo-27(tm5336) embryos expressing HIS-24::GFP
and GFP::RAB-7. Arrowheads indicate phagolysosomes positive for GFP::RAB-7.
The time point just prior to the appearance of GFP::RAB-7 on phagosomes was
set as “0 min”. Quantification is shown in e, with average duration (±SD) shown
in parentheses. 40 cell corpses were analyzed in each strain. f-i and q-w,
Confocal fluorescence images of the adult hypodermis in the indicated strains
expressing NUC-1::CHERRY without or with expression of DYN-1(WT)::GFP or
DYN-1(T67A)::GFP. Arrowheads indicate tubular lysosomes. Quantifications are
shown in j (left to right: n = 20, 13, 13, 15, 13 worms), k (left to right: n = 12, 10, 15,
10 worms), x (left to right: n = 18, 17, 21 worms), y (left to right: n = 18, 16, 17, 15
worms). l-m”, 3D reconstruction images of lysosomes in WT and dyn-1(ky51)
adults co-expressing HPO-27::GFP and NUC-1::CHERRY. Arrowheads indicate
association of HPO-27 with lysosomes. Quantification is shown in n (n = 15
worms per strain). o, Confocal fluorescence image of the hypodermis in a WT
adult co-expressing DYN-1::GFP and NUC-1::CHERRY. The boxed region is
magnified in the zoom image. White and magenta arrowheads indicate a DYN-1
punctum and a lysosome, respectively. Quantification is shown in p (n = 10 per
test). NUC-1::CHERRY and DYN-1::GFP puncta were analyzed separately to
yield Manders overlap coefficient M1 (DYN-1/NUC-1) and M2 (NUC-1/DYN-1),
respectively. The experiments were performed at 25 °C in (f-m”), which is
the non-permissive temperature of the dyn-1(ky51) allele. In c, j, k, n, p, x, y,
data are shown as mean ± SD. One-way ANOVA with Tukey’s multiple comparison
test (k, x, y) or Student’s two-tailed unpaired t-test (c, e, j, n) was performed to
compare mutant datasets with WT (c, e, j, k, n), datasets without and with
DYN-1(WT) or DYN-1(T67A) expression (x), all other datasets with hpo-27
RNAi (y) or datasets that are linked by lines. P values are indicated. Scale bars:
2.5 µm in (d), 5 µm in (a-b’, f-i, l-m”, o, q-w).
Extended Data Fig. 7 | HPO-27 is widely expressed and associates with
lysosomes. a-i’, DIC and confocal fluorescence images of WT animals at
different developmental stages carrying the HPO-27::mKate2 knock-in reporter.
The muscle cell, vulva, and spermatheca are indicated by green asterisks.
j-j”, Confocal fluorescence images of the hypodermis in WT expressing HPO-
27::mKate2 and Psemo-1 GFP. HPO-27::mKate2 puncta are present in the hyp7 cell
(GFP-positive; white arrowheads) and in the lateral seam cell (GFP-negative;
yellow arrowhead). k-l”, 3D reconstruction images of lysosomes in the
hypodermis of a WT adult expressing both HPO-27::mKate2 and SCAV-3::GFP
(k-k”’) or both MROH1::GFP and NUC-1::CHERRY (l-l”). The boxed region is
magnified in k”’. Arrowheads indicate association of HPO-27 or MROH1 with
lysosomes. m, n, Confocal fluorescence images of the hypodermis in WT adults
co-expressing NUC-1::CHERRY and HPO-27(ΔN)::GFP or HPO-27(ΔC)::GFP. Both
HPO-27 truncations fail to associate with lysosomes. In a-n, 15 animals were
examined in each strain. Scale bars: 5 µm.
Article

Extended Data Fig. 8 | See next page for caption.

Extended Data Fig. 8 | Loss of MROH1 affects acidity and proteolytic
activity of lysosomes. a-a”, Super-resolution images of a WT COS7 cell stably
expressing GFP-MROH1 and containing chased fluorescent Dextran A555. The
boxed region is magnified in the zoom images. Arrows and arrowheads indicate
association of MROH1 with vesicular and tubular lysosomes labeled by Dextran.
b, CRISPR-mediated generation of the MROH1 knockout COS7 cell line, and
verification by sequencing. The knockout clone contains a 28-bp deletion,
which causes a frameshift and results in a premature stop codon. c, qPCR
analysis of MROH1 expression in the indicated cells. Three independent
experiments were performed. Data are shown as mean±SD. d-o, Confocal
fluorescence images of WT (d-f), MROH1 KO (g-i), siControl ( j-l), and siMROH1
(m-o) COS7 cells stained by Lysotracker Red (d, g, j, m), Magic Red (e, h, k, n), or
DQ-BSA (f, i, l, o). Cell boundaries are outlined by dashed circles. Quantifications
are shown in p (n = 20 cells in each of 4 biological repeats), q (n = 20 cells in each
of 3 biological repeats). Data are shown as mean±SD. r-s”’, Confocal fluorescence
images of Lysotracker Blue-stained COS7 cells co-expressing GFP-MROH1 and
CHERRY-RAB7A(WT) (r-r”’) or CHERRY-RAB7A(T22N) (s-s”’). Arrowheads
indicate enrichment of MROH1 on RAB7-positive vesicles. In c, p, q, Student’s
two-tailed unpaired t-test was performed to compare MROH1 mutant or siRNA
datasets with WT or siControl. P values are indicated. Scale bars: represent 5 μm.
Article
Extended Data Fig. 9 | HPO-27 and MROH1 interact with RAB-7 and self-
interact in a “head-to-tail” mode. a, Schematic illustration of the tripartite
split-GFP assay. b-g’, DIC and confocal fluorescence images of the hypodermis
in WT adults co-expressing GFP1-9, NUC-1::CHERRY, WT or mutant forms of
GFP10::RAB-7, and HPO27::GFP11 (b-d’) or MROH1::GFP11 (e-g’). Arrowheads
indicate co-localization of reconstituted GFP and lysosomes. h-s’ DIC and
confocal fluorescence images of the hypodermis in WT adults co-expressing
GFP1-9, HPO-27::mKate2 (h-k’, p-s’) or NUC-1::CHERRY (l-o’), and the indicated
HPO-27 (h-k’), MROH1 (l-o’), HPO-27(ΔN) (p-q’), or HPO-27(ΔC) (r-s’) fusion
proteins. Arrowheads indicate co-localization of reconstituted GFP with HPO-
27::mKate2 or NUC-1::CHERRY. In b-s’, 15 animals were examined in each strain.
Scale bars: 5 µm.
Extended Data Fig. 10 | See next page for caption.
Article
Extended Data Fig. 10 | HPO-27 tends to oligomerize. a, Coomassie blue
staining of indicated purified recombinant proteins. Data are representative of
six independent experiments. b-e, Confocal fluorescence time-lapse images of
supported membrane tubes (SMrTs) labeled by Texas red DHPE without (b) or
with recombinant EGFP (c), HPO-27-EGFP (d) or MROH1-EGFP (e) proteins. “-PA”
indicates absence of phosphatidic acid (PA) in Texas red-labeled SMrTs. Three
independent experiments were performed. f-h, Tertiary structures of HPO-27
and human MROH1 predicated by RoseTTAFold. Comparison and illustrations
were prepared using the program PyMOL (https://pymol.org). Both proteins
contain 37 HEAT motifs, each of which is composed of two α-helices (A- and B-
helices) colored in green and brown, respectively. The first (1st) and the last
(37th) HEAT motifs are removed in HPO-27(ΔN) and HPO-27(ΔC), respectively.
Superimposition of the predicated tertiary structures of HPO-27 (green) and
MROH1 (brown) is shown in (g). Negatively charged (red) and positively charged
(blue) areas on the surface of an HPO-27 molecule are shown in a top (upper) or
bottom (lower) view in (h). i, Negative-staining EM showing monomeric and
oligomeric states of HPO-27. Corresponding cartoon illustrations are shown
underneath. j-r, Photo-bleaching and single-molecule imaging assay of
EGFP ( j-l), HPO-27-EGFP (m-o) and HPO-27(ΔC)-EGFP (p-r). TIRF images are
shown in (j, m, p). Representative traces show one-step photobleaching of EGFP
and HPO-27(ΔC)-EGFP, and two-step photobleaching of HPO-27-EGFP (arrows;
k, n, q). A summary of the number of photobleaching steps is shown in (l, o, r;
n = 200 foci in each of 3 biological repeats). Data are shown as mean ± SD. Scale
bars represent 5 µm in (b-e, j, m, p) and 20 nm in (i). s, Proposed model of HPO-27
function in lysosomal fission. For gel source data, see Supplementary Fig. 1.
 nature portfolio | reporting summary
Corresponding author(s): Xiaochen Wang, Wei Feng
Last updated by author(s): Feb 21, 2024

Reporting Summary
Nature Portfolio wishes to improve the reproducibility of the work that we publish. This form provides structure for consistency and transparency
in reporting. For further information on Nature Portfolio policies, see our Editorial Policies and the Editorial Policy Checklist.

Statistics
For all statistical analyses, confirm that the following items are present in the figure legend, table legend, main text, or Methods section.
n/a Confirmed
The exact sample size (n) for each experimental group/condition, given as a discrete number and unit of measurement
A statement on whether measurements were taken from distinct samples or whether the same sample was measured repeatedly
The statistical test(s) used AND whether they are one- or two-sided
Only common tests should be described solely by name; describe more complex techniques in the Methods section.

A description of all covariates tested

A description of any assumptions or corrections, such as tests of normality and adjustment for multiple comparisons
A full description of the statistical parameters including central tendency (e.g. means) or other basic estimates (e.g. regression coefficient)
AND variation (e.g. standard deviation) or associated estimates of uncertainty (e.g. confidence intervals)

For null hypothesis testing, the test statistic (e.g. F, t, r) with confidence intervals, effect sizes, degrees of freedom and P value noted
Give P values as exact values whenever suitable.

For Bayesian analysis, information on the choice of priors and Markov chain Monte Carlo settings
For hierarchical and complex designs, identification of the appropriate level for tests and full reporting of outcomes
Estimates of effect sizes (e.g. Cohen's d, Pearson's r), indicating how they were calculated
Our web collection on statistics for biologists contains articles on many of the points above.

Software and code

Policy information about availability of computer code
Data collection Image collection: Confocal laser scanning microscope (Zeiss LSM 880); Spinning disk confocal microscope(PerkinElmer); DeltaVision OMX V3
(3D-SIM) microscope (Cytiva, GE Healthcare); HT7800 (HITACHI); JEM-1400 (JEOL); Tecnai Spirit electron microscope (FEI); Multi-SIM
microscopy (Nanoinsights Multi- SIM); Axio Imager M2 microscope (Carl Zeiss); FCS2 system (Bioptechs)
qRT-PCR: Applied Biosystems QuantStudio 3 Flex real-time PCR system (Life Technologies)
Size-exclusion chromatography coupled with multi-angle light scattering (SEC-MALS): DAWN HELEOS II (Wyatt Technology); Optilab rEx (Wyatt
Technology); ASTRA (Wyatt Technology)

Data analysis Image processing: ZEN Imaging software (version 2.3); ImageJ (version 2.1.4.7); Volocity software (version 6.3); RELION (version 3.1.3); WoRx
(version 6.1.1); MATLAB (version R2017b); PyMOL (version 2.3.1)
Statistical analyses: GraphPad Prism (version 8.2.1)
For manuscripts utilizing custom algorithms or software that are central to the research but not yet described in published literature, software must be made available to editors and
reviewers. We strongly encourage code deposition in a community repository (e.g. GitHub). See the Nature Portfolio guidelines for submitting code & software for further information.
April 2023

1
 nature portfolio | reporting summary
Data
Policy information about availability of data
All manuscripts must include a data availability statement. This statement should provide the following information, where applicable:
- Accession codes, unique identifiers, or web links for publicly available datasets
- A description of any restrictions on data availability
- For clinical datasets or third party data, please ensure that the statement adheres to our policy

All data are available in the main text and supplementary materials. Structural prediction was performed with RoseTTAFold (http:://robetta.bakerlab.org) and
illustrations were prepared using the program PyMOL (http:://pymol.org). Numerical source data underlying all graphical representations and statistical descriptions
are provided in source data files with this paper. Full version of all gels and blots are provided in SI Fig. 1.

Research involving human participants, their data, or biological material

Policy information about studies with human participants or human data. See also policy information about sex, gender (identity/presentation),
and sexual orientation and race, ethnicity and racism.
Reporting on sex and gender N/A

Reporting on race, ethnicity, or N/A

other socially relevant
groupings

Population characteristics N/A

Recruitment N/A

Ethics oversight N/A

Note that full information on the approval of the study protocol must also be provided in the manuscript.

Field-specific reporting
Please select the one below that is the best fit for your research. If you are not sure, read the appropriate sections before making your selection.

Life sciences Behavioural & social sciences Ecological, evolutionary & environmental sciences
For a reference copy of the document with all sections, see nature.com/documents/nr-reporting-summary-flat.pdf

Life sciences study design

All studies must disclose on these points even when the disclosure is negative.
Sample size We determined sample size based on our previous publication(PMID: 32482227). For each experiment, the sample size is provided in the
legends and materials.

Data exclusions No data were excluded from any of the experimental studies presented in the manuscript.

Replication All experiments in this study were biologically repeated at least three independently and all attempts at replication were successful.

Randomization For all experiments, samples/worms/cells were randomly allocated into groups.

Blinding The data collection and analysis were blinded to group allocation.

Reporting for specific materials, systems and methods

We require information from authors about some types of materials, experimental systems and methods used in many studies. Here, indicate whether each material,
system or method listed is relevant to your study. If you are not sure if a list item applies to your research, read the appropriate section before selecting a response.
April 2023

2
Materials & experimental systems Methods

nature portfolio | reporting summary

n/a Involved in the study n/a Involved in the study
Antibodies ChIP-seq
Eukaryotic cell lines Flow cytometry
Palaeontology and archaeology MRI-based neuroimaging
Animals and other organisms
Clinical data
Dual use research of concern
Plants

Antibodies
Antibodies used Primary antibodies:
Mouse monoclonal anti-Flag (F3165, Sigma), Rabbit polyclonal anti-GFP (ab290, Abcam), Mouse monoclonal anti-MBP (E8032, NEB),
Mouse monoclonal anti-HPO-27 (Antibody Core, NIBS, this paper), Rat polyclonal anti-ASP-4(Antibody Core, NIBS, this paper), Rabbit
polyclonal anti-CHERRY (26765-1-AP, Proteintech), Rabbit polyclonal anti-γ-tubulin(T5192, Sigma)
Secondary antibodies:
Goat Anti-Mouse IgG (H+L) (115-035-003, Jackson), Goat Anti-Rabbit IgG (H+L) (111-005-003, Jackson), Goat Anti-Rat IgG (H+L)
(112-005-003, Jackson)

Validation 1. Mouse monoclonal anti-Flag (F3165, Sigma): Validated by the supplier with the following notes:(1) Species reactivity: all. (2)
Application: immunoblotting, immunoprecipitation, immunocytochemistry, immunofluorescence, ELISA, EIA, chromatin
immunoprecipitation, electron microscopy, flow cytometry, supershift assays.
2. Rabbit polyclonal anti-GFP (ab290, Abcam): Validated by the supplier with the following notes:(1) Species reactivity: all. (2)
Application: ELISA, IHC-Fr, ICC, IHC-P, IP, WB, IHC-FoFr, IHC-FrFl, Electron Microscopy.
3. Mouse monoclonal anti-MBP (E8032, NEB): Validated by the supplier with the following notes:(1) Species reactivity: all. (2)
Application: ELISA, WB.
4. Mouse monoclonal anti-HPO-27 (Antibody Core, NIBS, this paper): Validated by the supplier with the following notes:(1) Species
reactivity: worm. (2) Application: WB.
5. Rat polyclonal anti-ASP-4(Antibody Core, NIBS, this paper): Validated by the supplier with the following notes:(1) Species reactivity:
worm. (2) Application: WB.
6. Rabbit polyclonal anti-CHERRY (26765-1-AP, Proteintech): Validated by the supplier with the following notes:(1) Species reactivity:
all. (2) Application: IF, WB, ELISA.
7. Rabbit polyclonal anti-γ-tubulin(T5192, Sigma): Validated by the supplier with the following notes: (1) Species reactivity: human,
chicken. (2) Application: immunocytochemistry, immunofluorescence, immunoblotting, staining in microscopy, chromatin
immunoprecipitation (ChIP).

Eukaryotic cell lines

Policy information about cell lines and Sex and Gender in Research
Cell line source(s) COS7 (CRL-1651, ATCC), sf9 (11496015, gibco)

Authentication The cell line was authenticated by STR analysis by in this study.

Mycoplasma contamination The cell line was validated to be free of mycoplasma contamination before used in this study.

Commonly misidentified lines No commonly misidentified cell lines were used.

(See ICLAC register)

Animals and other research organisms

Policy information about studies involving animals; ARRIVE guidelines recommended for reporting animal research, and Sex and Gender in
Research

Laboratory animals
In the article, we used nematodes at the embryonic stage (comma, 1.5 fold, 2 fold, 2.5 fold, 3 fold, 4 fold), larval stage [(L1,
L4(intermolt, molt)] and adult stage (day 1, day 2, day 3, day 4, day 5).
Caenorhabditis elegans strain:
hpo-27(tm5336) LG I, hpo-27(qx427) LG I, hpo-27(qx428) LG I, hpo-27(qx429) LG I, hpo-27(qx431) LG I, hpo-27(qx432) LG I,
hpo-27(qx436) LG I, hpo-27(qx437) LG I, hpo-27(qx458) LG I, eat-2(ad1116) LG II, epg-5(tm3425) LG II, cup-5(bp510) LG III,
daf-2(e1370) LG III, isp-1(qm150) LG IV, ppk-3(n2668) LG X, dyn-1(ky51) LG X, qxIs257(ced-1p::NUC-1::CHERRY),
qxIs569(semo-1p::HPO-27::GFP), qxIs544(rnst-2p::RNST-2::CHERRY), qxIs779(scav-3p::SCAV-3::tagRFPt),
April 2023

qxIs430(scav-3p::SCAV-3::GFP), qxIs68(ced-1p::CHERRY::RAB-7), qxIs468(myo-3p::LAAT-1::GFP), qxIs439(semo-1p::GFP::TRAM-1),

qxIs619(semo-1p::HPO-27::Flag), qxIs377(semo-1p::DYN-1::GFP), yqIs157(semo-1p::Mito-GFP), qxCas88(HPO-27::TurboID::mKate2),
qxSi13(lgg-1p::GFP::LGG-1), qxSi36(scav-3p::SCAV-3::RFP::3xHA), kuIs47(ajm-1p::AJM-1::GFP), zuIs45(nmy-2p::NMY-2::GFP),
rmIs132(unc-54p::Q35::YFP), qxIs864(col-12p::LMP-1::RFP), qxIs734(scav-3p::SCAV-3::APEX::sfGFP), qxIs66(ced-1p::GFP::RAB-7),
smIs18(HIS-24::GFP), qxEx9593(semo-1p::GFP), qxEx7806[semo-1p::RAB-7(T23N)], qxEx7824[semo-1p::RAB-7(Q68L)],
qxEx5479[Fosimid(WRM0613cD08)], qxEx9285[semo-1p::HPO-27(del C)::GFP], qxEx9398[semo-1p::HPO-27(del N)::GFP],

3
 qxEx8775(semo-1p::MROH1::GFP), qxEx4274(vha-6p::LAAT-1::GFP), qxEx9400(vha-6p::HPO-27::GFP),
qxEx9587[semo-1p::GFP1-9, semo-1p::GFP10::RAB-7(WT), semo-1p::HPO-27::GFP11],

nature portfolio | reporting summary

qxEx9589[semo-1p::GFP1-9, semo-1p::GFP10::RAB-7(T23N), semo-1p::HPO-27::GFP11],
qxEx9588[semo-1p::GFP1-9, semo-1p::GFP10::RAB-7(Q68L), semo-1p::HPO-27::GFP11],
qxEx9590[semo-1p::GFP1-9, semo-1p::GFP10::RAB-7(WT), semo-1p::MROH1::GFP11],
qxEx9592[semo-1p::GFP1-9, semo-1p::GFP10::RAB-7(T23N), semo-1p::MROH1::GFP11],
qxEx9591[semo-1p::GFP1-9, semo-1p::GFP10::RAB-7(Q68L), semo-1p::MROH1::GFP11],
qxEx9264(semo-1p::GFP1-9, semo-1p::GFP10::HPO-27, semo-1p::GFP11::HPO-27),
qxEx9404(semo-1p::GFP1-9, semo-1p::HPO-27::GFP10, semo-1p::HPO-27::GFP11),
qxEx9405(semo-1p::GFP1-9, semo-1p::GFP10::HPO-27, semo-1p::HPO-27::GFP11),
qxEx9278(semo-1p::GFP1-9, semo-1p::HPO-27::GFP10, semo-1p::GFP11::HPO-27),
qxEx9406[semo-1p::GFP1-9, semo-1p::GFP10::HPO-27(del N), semo-1p::HPO-27(del N)::GFP11],
qxEx9407[semo-1p::GFP1-9, semo-1p::HPO-27(del N)::GFP10, semo-1p::GFP11::HPO-27(del N)],
qxEx9408[semo-1p::GFP1-9, semo-1p::GFP10::HPO-27(del C), semo-1p::HPO-27(del C)::GFP11],
qxEx9409[semo-1p::GFP1-9, semo-1p::HPO-27(del C)::GFP10, semo-1p::GFP11::HPO-27(del C)],
qxEx9410(semo-1p::GFP1-9, semo-1p::GFP10::MROH1, semo-1p::GFP11::MROH1),
qxEx9411(semo-1p::GFP1-9, semo-1p::MROH1::GFP10, semo-1p::MROH1::GFP11),
qxEx9412(semo-1p::GFP1-9, semo-1p::GFP10::MROH1, semo-1p::MROH1::GFP11),
qxEx9413(semo-1p::GFP1-9, semo-1p::MROH1::GFP10, semo-1p::GFP11::MROH1),
qxEx9684[semo-1p::DYN-1(T67A)::GFP]

Wild animals No wild animals were used in this study.

Reporting on sex Not applicable.

Field-collected samples This study did not involve field collected samples.

Ethics oversight No ethical approval was required.

Note that full information on the approval of the study protocol must also be provided in the manuscript.

Plants
Seed stocks Report on the source of all seed stocks or other plant material used. If applicable, state the seed stock centre and catalogue number. If
plant specimens were collected from the field, describe the collection location, date and sampling procedures.

Novel plant genotypes Describe the methods by which all novel plant genotypes were produced. This includes those generated by transgenic approaches,
gene editing, chemical/radiation-based mutagenesis and hybridization. For transgenic lines, describe the transformation method, the
number of independent lines analyzed and the generation upon which experiments were performed. For gene-edited lines, describe
the editor used, the endogenous sequence targeted for editing, the targeting guide RNA sequence (if applicable) and how the editor
was applied.
Authentication Describe any authentication procedures for each seed stock used or novel genotype generated. Describe any experiments used to
assess the effect of a mutation and, where applicable, how potential secondary effects (e.g. second site T-DNA insertions, mosiacism,
off-target gene editing) were examined.

April 2023