Methods in cell biology, volume 130

Cell surface turnover in these cells is rapid and vol-uminous and so REs also receive membrane from the cell surface through constantmacropinocytic and, sometimes, phagocytic pathways, w

Biology Sorting and Recycling

Endosomes

Volume 130

Philadelphia, USA &

Institut Curie, Paris, France

Biology Sorting and Recycling

Endosomes

Volume 130

Edited by

Wei Guo

University of Pennsylvania, Biology Department,

Philadelphia, PA, USA

AMSTERDAM • BOSTON • HEIDELBERG • LONDON

NEW YORK • OXFORD • PARIS • SAN DIEGO

SAN FRANCISCO • SINGAPORE • SYDNEY • TOKYO

Academic Press is an imprint of Elsevier

The Boulevard, Langford Lane, Kidlington, Oxford OX5 1GB, UK

First edition 2015

Copyright© 2015 Elsevier Inc All Rights Reserved

No part of this publication may be reproduced or transmitted in any form or by any means,electronic or mechanical, including photocopying, recording, or any information storageand retrieval system, without permission in writing from the publisher Details on how toseek permission, further information about the Publisher’s permissions policies and ourarrangements with organizations such as the Copyright Clearance Center and theCopyright Licensing Agency, can be found at our website:www.elsevier.com/permissions.This book and the individual contributions contained in it are protected under copyright bythe Publisher (other than as may be noted herein)

Notices

Knowledge and best practice in this field are constantly changing As new research andexperience broaden our understanding, changes in research methods, professionalpractices, or medical treatment may become necessary

Practitioners and researchers must always rely on their own experience and knowledge inevaluating and using any information, methods, compounds, or experiments describedherein In using such information or methods they should be mindful of their own safetyand the safety of others, including parties for whom they have a professional responsibility

To the fullest extent of the law, neither the Publisher nor the authors, contributors, oreditors, assume any liability for any injury and/or damage to persons or property as amatter of products liability, negligence or otherwise, or from any use or operation of anymethods, products, instructions, or ideas contained in the material herein

ISBN: 978-0-12-802829-2

ISSN: 0091-679X

For information on all Academic Press publications

visit our website athttp://store.elsevier.com

Gerard Apodaca

Departments of Medicine and Cell Biology, University of Pittsburgh, Pittsburgh,

PA, USA

Daniel D Billadeau

Department of Biochemistry and Molecular Biology, Division of Oncology

Research, Mayo Clinic, Rochester, MN, USA

Department of Biochemistry and Molecular Biology and the Fred and Pamela

Buffett Cancer Center, University of Nebraska Medical Center, Omaha, NE, USA

Pei-Wen Chen

Laboratory of Cellular and Molecular Biology, Center for Cancer Research,

National Cancer Institute, Bethesda, MD, USA

Xiao-Wei Chen

Institute of Molecular Medicine, Peking University, Beijing, China; PKU-THU

Center for Life Sciences, Peking University, Beijing, China

Nicholas D Condon

Institute for Molecular Bioscience, The University of Queensland, Brisbane, QLD,

Australia

Paul de Figueiredo

Department of Molecular Biology and Genetics, Cornell University, Ithaca, NY,

USA; Department of Microbial Pathogenesis and Immunology, Norman Borlaug

Center, Texas A&M University, College Station, TX, USA

Ce´dric Delevoye

Institut Curie, PSL Research University, Paris, France; CNRS UMR 144,

Structure and Membrane Compartments, Paris, France

Emmanuel Derivery

Department of Biochemistry, Sciences II, University of Geneva, Geneva,

Switzerland

xiii

Julie G Donaldson

Cell Biology and Physiology Center, National Heart, Lung, and Blood Institute,National Institutes of Health, Bethesda, MD, USA

Michael Robert Dores

Department of Pharmacology, School of Medicine, University of California,San Diego, La Jolla, CA, USA

Heidi Hehnly

Program in Molecular Medicine, University of Massachusetts Medical School,

Worcester, MA, USA; Department of Cell and Developmental Biology, State

University of New York Upstate Medical University, Syracuse, NY, USA;

Department of Pharmacology, University of Washington, Seattle, WA, USA

Victor W Hsu

Division of Rheumatology, Immunology, and Allergy, Brigham and Women’s

Hospital, Boston, MA, USA; Department of Medicine, Harvard Medical School,

Boston, MA, USA

Hui-Fang Hung

Program in Molecular Medicine, University of Massachusetts Medical School,

Worcester, MA, USA

Xiaoying Jian

Laboratory of Cellular and Molecular Biology, Center for Cancer Research,

National Cancer Institute, Bethesda, MD, USA

Danielle N Kalkofen

Department of Molecular Biology and Genetics, Cornell University, Ithaca, NY,

USA

Guangpu Li

Key Laboratory of Biopesticide and Chemical Biology, Fujian Agriculture

and Forestry University, Fuzhou, China; Department of Biochemistry and

Molecular Biology, University of Oklahoma Health Sciences Center, Oklahoma

City, OK, USA

Jian Li

Division of Rheumatology, Immunology, and Allergy, Brigham and Women’s

Hospital, Boston, MA, USA; Department of Medicine, Harvard Medical School,

Boston, MA, USA

Zhimin Liang

Department of Biochemistry and Molecular Biology, University of Oklahoma

Health Sciences Center, Oklahoma City, OK, USA

Guodong Lu

Key Laboratory of Biopesticide and Chemical Biology, Fujian Agriculture and

Forestry University, Fuzhou, China

Ruibai Luo

Laboratory of Cellular and Molecular Biology, Center for Cancer Research,

National Cancer Institute, Bethesda, MD, USA

Andres E Perez Bay

Margaret Dyson Vision Research Institute, Weill Medical College of CornellUniversity, New York, NY, USA

Brian C Richardson

Weill Institute for Cell and Molecular Biology, Cornell University, Ithaca, NY, USA

Enrique Rodriguez-Boulan

Margaret Dyson Vision Research Institute, Weill Medical College of Cornell

University, New York, NY, USA

Ryan Schreiner

Margaret Dyson Vision Research Institute, Weill Medical College of Cornell

University, New York, NY, USA

Anbing Shi

Department of Medical Genetics, School of Basic Medicine and the Collaborative

Innovation Center for Brain Science, Tongji Medical College, Huazhong

University of Science and Technology, Wuhan, Hubei, China; Institute for Brain

Research, Huazhong University of Science and Technology, Wuhan, Hubei,

Department of Cell Biology, University of Pittsburgh School of Medicine,

Pittsburgh, PA, USA

Jennifer L Stow

Institute for Molecular Bioscience, The University of Queensland, Brisbane, QLD,

Australia

JoAnn Trejo

Department of Pharmacology, School of Medicine, University of California,

San Diego, La Jolla, CA, USA

Key Laboratory of Biopesticide and Chemical Biology, Fujian Agriculture and

Forestry University, Fuzhou, China

Bin Wu

Department of Biology, University of Pennsylvania, Philadelphia, PA, USA

Endosomes not only serve as a receiving compartment for proteins endocytosed

from cell surface, but also function as a donor compartment and sorting station

for cargos designated to lysosomes, Golgi, and plasma membrane In recent years,

the importance of endosomes as sorting and recycling compartments has become

increasingly appreciated However, the complexity of the endosome system also

di-vulges the gap in our knowledge of this fascinating system Like their sophisticated

cousin Golgi, endosomes are dynamic and functionally versatile, yet they are even

more elusive due to their highly elaborate and heterogeneous structures Scientists

from various fields of cell biology, membrane traffic, and beyond, see the need to

communicate the methods used to study the dynamics and functions of this

endo-membrane system

The aim of this volume is to bring together specialists from the field to contribute

their expertise to a broad range of biomedical researchers interested in endosomal

sorting and recycling The first part of the volume consists of chapters focusing

on the Rab family of small GTPases, which are not only key regulators of different

trafficking steps, but also landmarks of various endosomal compartments Methods

for biochemical, imaging, and functional studies of Rab proteins and their regulators

are included (Chapters 1e5) These chapters are followed by approaches studying

the Arf GTPases as they control the generation of tubular vesicular carriers from

the endosomes (Chapters 6e9) In addition, methods to study the proteins and lipids

that control membrane morphology and dynamics are presented (Chapters 10e12)

As in most cases, vesicle budding and transport from endosomes involve the action

of, and coordination with, cytoskeleton Several chapters present methods to study

the actin polymerization machinery and microtubule-based transport during

endoso-mal trafficking (Chapters 13e15) This volume also includes studies of epithelial

cells, which are characterized by their complicated sorting machinery that ensures

the correct targeting of proteins to distinct apical and basolateral domains at the

plasma membrane Experimental protocols are presented to investigate the various

routes of protein transport from endosomes in this polarized system (Chapters

16e18) The last part of the volume presents studies that link endosomal trafficking

and signaling (Chapters 19e22) Elucidating the mechanisms of receptor recycling

and degradation is essential for the understanding of the endosome function as

well as signal transduction Biochemical and microscopic imaging approaches

are presented for quantitative studies of receptor internalization, recycling, and

degradation

Clearly, due to the complexity of the system and rapid progress of the field, these

chapters cannot cover all aspects of endosomal sorting and recycling It is my hope

that the expertise gathered in this volume will serve as a useful resource and platform

for future investigation of this fascinating membrane system

Wei Guo

xix

Dynamic imaging of the

recycling endosomal

Adam A Wall * , Nicholas D Condon * , Jeremy C Yeo * ,

Nicholas A Hamilton * ,x, Jennifer L Stow * , 1

*Institute for Molecular Bioscience, The University of Queensland, Brisbane, QLD, Australia

xResearch Computing Centre, The University of Queensland, Brisbane, QLD, Australia

1

Corresponding author: E-mail: j.stow@imb.uq.edu.au

CHAPTER OUTLINE

Introduction 2

1 Methods 4

1.1 Cell Culture and Transfection 4

1.2 Materials and Instruments 4

1.2.1 Reagents 4

1.2.2 Instruments 5

1.3 Live Cell Imaging 5

2 Experimental Strategies 5

2.1 Strategy 1: RE Subcompartments Defined by Rabs 5

2.1.1 Background and objective 5

2.1.2 Flow of the experiment 6

2.1.3 Results 6

2.2 Strategy 2: Cargo Movement through RE 8

2.2.1 Background and Objective 8

2.2.2 Flow of the experiments 9

2.2.3 Experimental considerations 11

2.2.4 Results 11

2.3 Strategy 3: Delivery of RE to Phagosomes 14

2.3.1 Background and objective 14

2.3.2 Flow of the experiment 15

2.3.3 Experimental considerations 15

2.3.4 Results 16

Summary 16

References 17

Methods in Cell Biology, Volume 130, ISSN 0091-679X, http://dx.doi.org/10.1016/bs.mcb.2015.04.007

Recycling endosomes (REs) form an extensive and complex network of compartmentalized vesicular and tubular elements that connect with the cell surface andother endosomes in macrophages As surveillance and defense cells of the innate immunesystem, macrophages are highly dependent on REs for their active and voluminous cellsurface turnover and endocytic, exocytic, and recycling of membrane and cargo Here weset out three approaches for imaging and analyzing REs in macrophages, based on theexpression of fluorescently labeled RE-associated proteins and the uptake of fluorescentcargo Subcompartments of the REs are identified by co-expression and co-localizationanalysis of RE associated Rab GTPases Transferrin is a well-known cargo marker as itrecycles through REs and it is compared here to other cargo, revealing how differentendocytic routes intersect with REs We show how the movement of transferrin throughREs can be modeled and quantified in live cells Finally, since phagosomes are a signatureorganelle for macrophages, and REs fuse with the maturing phagosome, we show imaging

sub-of REs with phagosomes using a genetically encoded pH-sensitive SNARE-based probe.Together these approaches provide multiple ways to comprehensively analyze REs and theimportant roles they play in these immune cells and more broadly in other cell types

INTRODUCTION

In most eukaryotic cells, recycling endosomes (REs) form a central hub necting endocytic and exocytic trafficking (Hsu & Prekeris, 2010; Maxfield &McGraw, 2004) In macrophages, the REs form a vast, interconnected network ofvesicular and tubular elements that are collectively distributed from the perinuclearregion to the subplasma membrane zone (Lacy & Stow, 2011; Manderson, Kay,Hammond, Brown, & Stow, 2007) This highly dynamic organellar system is crucialfor the immune surveillance roles of macrophages, which include the ingestion andsampling of fluid, particulate matter and pathogens, the destruction of pathogens andcell debris, and the secretion of critical mediators like chemokines and cytokines(Epelman, Lavine, & Randolph, 2014) REs are in intimate contact with the plasmamembrane, providing a reservoir of extra membrane for the formation of cell surfaceprojections, like filopodia, lamellipodia, dorsal ruffles, and phagocytic cups (Huynh,Kay, Stow, & Grinstein, 2007) Cell surface turnover in these cells is rapid and vol-uminous and so REs also receive membrane from the cell surface through constantmacropinocytic and, sometimes, phagocytic pathways, which feed membrane backinto the intracellular recycling pool (Hsu & Prekeris, 2010) Finally, REs are alsosecretory compartments, acting as a sorting substation for many recycling or newlysynthesized cargo proteins and then fusing with the plasma membrane to deliver pro-teins to the cell surface or releasing secreted proteins (Lacy & Stow, 2011; Murray,Kay, Sangermani, & Stow, 2005) Thus, REs traffic a wide variety of cargo and largeexpanses of membrane into and out of the cell They also participate in receptorsignaling pathways for cell activation via pathogen-associated and danger signals,and opsins such as immunoglobulin and complement (Husebye et al., 2010; Nair-Gupta et al., 2014) Growth factor receptors and G protein-coupled receptors often

intercon-traverse the RE network in routes that help to determine the fate of signaling through

receptor recycling or degradation (Marchese, 2014)

The highly dynamic nature of REs in macrophages means that this compartment

is optimally visualized and its functions best appreciated through fluorescence

imaging in live cells Labeled REs can be readily imaged in fixed macrophages as

a series of small puncta, contrasting with imaging in live cells where the dynamic

connections and tubules that interconnect the RE network and provide connections

to other parts of the cell, can be resolved (Manderson et al., 2007) The expression of

fluorescently tagged proteins in primary macrophages and macrophage cell lines has

revealed many of the unique structure/function aspects of REs in these cells The

combined use of fluorescently tagged resident RE cellular proteins and labeled

cargo, allows for visualization and quantification of endocytic and exocytic traffic

through REs Coexpression of multiple, tagged RE-associated proteins shows the

complex subcompartmentalized nature of the RE network itself

Members of the Rab family of small GTPases have proven to be invaluable

exper-imental markers for REs and other organelles (Chavrier, Parton, Hauri, Simons, &

Zerial, 1990; Hutagalung & Novick, 2011) In their active, GTP-bound form, Rabs

bind to specific organelles or membrane carriers to participate in membrane

traf-ficking, fusion, budding, and in signalling (Das & Guo, 2011; Hutagalung & Novick,

2011) The quintessential RE marker across many cell types is Rab11a and, in

mac-rophages too, this protein defines a major subpopulation of REs, particularly those

containing cargo destined for the cell surface (Das & Guo, 2011; Murray et al.,

2005; Stow & Murray, 2013)

Members of the SNARE protein family are also commonly used for demarking

membrane compartments in cells (Stow, Manderson, & Murray, 2006) SNAREs

2009; Su¨dhof & Rothman, 2009; Wickner, 2010) and as such, the majority of

SNAREs are transmembrane (TM) proteins with long cytoplasmic tails containing

a coiled-coil SNARE domain and short lumenal or extracellular domain As

“resi-dent” proteins, SNAREs can be used as markers for endosomes, either by

immuno-labeling or through expression of GFP-tagged SNAREs In macrophages, VAMP3 is

well-established as an RE marker, where it functions by interacting with the

SNAREs syntaxin4 and SNAP-23 on the plasma membrane or on phagosomes to

fuse REs and their cargo to these sites (Nair-Gupta et al., 2014; Stow et al., 2006)

Basic methods for visualizing RE-associated proteins and lipids in live

macro-phages form the basis for studying a host of receptors, secreted and endocytosed

products that traverse REs Understanding RE behavior and connections to the

cell surface and to other organellesdparticularly phagosomes, autophagosomes,

lysosomes, Golgi, endoplasmic reticulum, and the cell surfacedis essential for

elucidating macrophage immune responses in homeostasis and in a variety of

infec-tions and inflammatory and developmental diseases Macrophages provide perhaps

one of the most extreme examples of REsdin terms of the volume, dynamism, and

variety of functionsdof any cell type For the same reasons, macrophages are ideal

subjects for imaging and analyzing REs

1 METHODS

Macrophages of the macrophage-like line RAW264.7 are maintained in RPMI plemented with 10% (v/v) fetal bovine serum (pre-screened for non-activation ofcells) and 1% (v/v) L-glutamine at 37C, 5% CO2 incubator Macrophages arepassaged in 10 cm non-tissue culture treated plates and split thrice weekly forexperimental use Fixed cell imaging requires macrophages to be seeded on #1,

sup-11 mm glass coverslips within a 24-well tissue culture plate at a density of0.1 106cells/mL for next day transfection or 0.2 106cells/mL for next day fix-ation Macrophages to be imaged live are seeded onto 35 mm MatTek glass bottomdishes at the same cell densities

Transient transfection of fluorescent DNA constructs by lipofection is performedapproximately 16 h after plating when cells are in exponential growth phase,yielding better expression in the often difficult-to-transfect macrophages Cellsare incubated prior to transfection with reduced serum medium (OptiMEM, Gibco).For transfection, we use 2 mg plasmid DNA per MatTek or 0.5 mg per coverslips in a24-well tissue culture plate following the recommended protocol for Lipofectamine

2000 (Life-Technologies) Cells are incubated with the DNA/Lipofectamine for 2 h

at 37C, 5% CO2, before being replaced with regular growth medium Macrophagescan be imaged from 6 to 16 h post transfection This methodology yields a typicaltransfection efficiency of 10e30% that is more than sufficient for imagingexperiments

Primary mouse macrophages are transfected for imaging using the Amaxa ofection method according to manufacturer’s protocol In brief, cells are washedwith OptiMEM and 1 106cells are used per reaction in a total of 100 mL volume.Cells are then resuspended in OptiMEM and mixed with 10 mg of DNA per Amaxaelectroporation cuvette Electroporation is carried out using the D-032 setting as permanufacturer’s protocol Cells are then immediately transferred to prewarmed com-plete RPMI medium for recovery

1.2.1 Reagents

RPMI culture medium

Fetal bovine serum

L-Glutamine

Lipofectamine2000

OptiMEM

30 mm Latex Beads

Lipopolysaccaharide, Salmonella minnesota R595 (LPS)

Fluorescently tagged cargo: Alexa647 transferrin, Alexa488-dextran

(10,000 mw), and Alexa594-low density lipoprotein (LDL)

1.2.2 Instruments

Applied precision personal DeltaVision microscope

Based on Olympus IX71 inverted wide-field epi-fluorescence and

bright-field microscope fitted with a Roper Coolsnap HQ2 monochrome camera

with a 120 W xenon lamp coupled with an incubation chamber and CO2

control

RAW264.7 macrophages are seeded onto a 35 mm MatTek glass bottom dishes For

next day imaging, a total of 0.8 106cells are plated or for next day transfection we

use a total of 0.4 106

cells to allow for cell growth and to maintain optimal ing conditions

imag-Where needed, macrophages are activated by adding LPS at 100 ng/mL to the

medium from 30 min before imaging Activated macrophages have more active

traf-ficking and more dynamic REs than preactivated cells

The Personal DeltaVision deconvolution microscope system is used for

time-lapse imaging for rapid, wide-field illumination and acquisition of images The

microscope is equipped with a chamber for maintaining live cells at 37C and

30 min prior to use for maximum stabilization of the imaging system If CO2

infu-sion is not available; CO2-independent medium can be used The 60 objective with

objective oil of 1.20 refractive index is typically used to image single cells In our

live cell examples, we utilize the DeltaVisions optical axis integration (OAI) scan

function to limit the time required to take a small optical z-scan This helps to speed

up image acquisition and decrease photobleaching and photo-toxicity Imaging

con-ditions are carefully selected to minimize photobleaching and pixel saturation

dur-ing image acquisition

2.1.1 Background and objective

The RE network is highly compartmentalized with subcompartments that collect

Marchese, 2014) Membranes that pinch off the RE network are also carriers that

can fuse with the plasma membrane to deliver exocytic or secretory cargo As in

epithelial cells, REs in macrophages organize “polarized” trafficking to different

cell surface projections or to the front and back of migrating cells (Das & Guo,

2011; Hsu & Prekeris, 2010) The multiple members of the Rab family associate

with and demark different subcompartments of REs Comparing two or more tagged

Rabs in cotransfected cells for instance highlights the segregation of Rabs often on

overlapping but also distinct RE membranes Rabs11a and 8a are sequentially actingRabs on REs in exocytic pathways (Das & Guo, 2011), while they often exhibitcoordinated actions through their effectors and accessory proteins, they typicallyoccupy nonidentical membrane domains in the RE network An important pointfor further investigation is how guanine nucleotide exchange factors and othercellular machinery target and retain Rabs on these very specific membrane domains.Since specific Rabs are perhaps the most recognizable markers for REs, express-ing and localizing fluorescently tagged Rabs is a mainstay of RE studies To high-light RE subcompartments in live cells, and to assign individual Rabs to differentstructural and functional aspects of the RE network in macrophages, we describethe coexpression and imaging of fluorescently tagged RE-associated GTPases,Rab8a, and Rab11a Imaging in fixed and live cells highlights the segregation ofthese two Rabs on different membrane domains within the RE network.

2.1.2 Flow of the experiment

1 Transfect macrophages plated on coverslips for fixed cells or MatTek dishes (for

live imaging) with a 1:1 mixture of GFP-Rabb1a and td-tomato-Rab8a

plasmids

2 Wash cells and replace with RPMI medium for 16 h.

3 Cells can be image fixed or live on a fluorescence microscope of choice (e.g.,

Personal DeltaVision deconvolution microscope)

4 Prior to experiments, cells are stimulated with 100 ng/mL of LPS for 30 min to

activate them

5 For fixed cell examples, images are merged using ImageJ to see overlap and

colocalization

6 For live cell examples, images are collected using time-lapse microscopy that is

defined by the fluorescent intensity of the transfected proteins Using the taVision microscope, we utilize the OAI scan function and keep the lamp power

Del-at 30% or bellow to limit photo-toxicity For fixed cell examples, a higher powercan be utilized

7 Aquired OAI scans are deconvolved as projections using Applied Precision’s

online microscope software and visualized using ImageJ

2.1.3 Results

In fixed macrophages, REs appear as puncta throughout the cytoplasm and in centrations in the perinuclear area (Figure 1(A)) GFP-Rab11a is on many puncta,consistent with its widespread distribution on REs Rab8a is much less abundant

con-on internal membranes It is found con-on some RE puncta but also con-on surface rufflesand some larger, circular macropinosomes or short tubular elements The overlap

of Rabs 8a and 11a can be seen on specific RE puncta (Figure 1(A), box) where amples demonstrate the range of colocalization scenarios for these two RE Rabs.Rabs 8a and 11a can be on separate RE puncta, they can be seen as adjacent labeling

ex-on the same puncta and ex-on a few REs these is complete overlap of the two labeled

FIGURE 1 Localization of recycling endosome-associated Rab GTPases in macrophages.

RAW264.7 macrophages coexpressing GFP-Rab11a and td-Tomato-Rab8a (A) Fixed cells show perinculear concentrations of active

Rabs11a and 8a on REs and scattered RE puncta throughout the cells Rab8a is additionally found on larger peripheral macropinosomes

The inset shows a selection of punctate REs highlighting examples of complete overlap of both Rabs, their side-by-side juxtaposition on

some REs and separately labeled REs, together representing the dynamic flux of membranes containing these Rabs on intersecting

subcompartments (B) Live cell imaging of cotransfected macrophages The REs in live cells appear much more tubular, often ringlike

and their dynamic nature can be appreciated in movies Frames from a movie show Rab8a positive tubular membranes forming

macropinosomes from surface ruffles and coursing back into the REs from the surface These tubules skirt around a vacuole and intersect

with Rab11a REs (arrowheads) Scale bar¼ 10 mm (See color plate)

Rabs These distributions reflect the presence of Rabs11a and 8a on highly dynamic,intersecting membranes within the RE network.

Live imaging provides a clearer picture of the dynamic nature of membranes labeledfor the two Rabs In the cell depicted (Figure 1(B)), GFP-Rab11a appears on punctatemembranes scattered throughout the RE network In movies, these Rab11a puncta can

be seen moving around in the cell, often moving toward the cell surface or movingback and forth between the cytoplasm and cell surface In contrast, td-tomato-Rab8a is found on surface ruffles in macrophages (Luo et al., 2014) and on verytubular membranes that typically form from the cell surface and bring membraneback into the REs Live cell imaging in macrophages highlights the prominent andperhaps preferential association of Rab11a with exocytic components of REs and

of Rab8a with membrane being retrieved from the cell surface back into REs

2.2.1 Background and Objective

Interconnected endosomal networks within macrophages handle many differenttypes of cargo molecules, with varied destinations (Maxfield & McGraw, 2004).Different cytokines traverse the REs during movement from the trans-Golgi networktowards the cell surface for secretion (Manderson et al., 2007; Stow et al., 2013;Beaumont et al., 2011) Receptors can be internalized into REs and then targetedfor degradation or recycling (Marchese, 2014) The RE is also important forhandling receptors during pathogen contact (Husebye et al., 2010) and for bringingnutrients into the cell on recycling surface receptors (Maxfield & McGraw, 2004).Transferrin receptors in complex with the iron chelating protein transferrin traversethis network, bringing iron into the cell and then returning to the surface for newrounds of uptake Both transferrin receptors and labeled transferrin are quintessen-tial markers of REs in most cell types (Maxfield & McGraw, 2004)

For studies on REs, fluorescently labeled cargo can be used to demark the ment itself and to track movement of the cargo and the REs through endocytic, recy-cling, or exocytic routes Tracking multiple cargos, such as different cytokines, canalso demonstrate how cargo is sorted and compartmentalized within the RE network(Manderson et al., 2007) Tracking and measuring the rates at which fluorescentlytagged cargo moves throughout the RE network can provide insights into the func-tions of RE-associated proteins and it can be used to measure disease-related celldysfunction Here we describe two experiments that can be used to study the function

compart-of RE associated proteins The first uses three different endocytic markers to identifythree trafficking routes that all intersect with the RE cargo transferrin at some pointduring recycling Fluorescent dextran labels all fluid phase entry into the cell such asmacropinocytosis, while LDL represents receptor-mediated entry Both that bth ulti-mately lead to a degradative fate in the lysosomes The second experiment measuresthe half-life of transferrin in the RE by measuring the transferrin total fluorescenceover-time These two experiments can be used to study the function of RE residentproteins in both knockdown or over-expression studies

2.2.2 Flow of the experiments

Experiment 1: Recycling endosome component trafficking pathways

Stage #1 Experimental Setup and Acquisition

1 RAW264.7 macrophages are seeded onto #1.5 glass coverslips the day prior

to experimentation

2 The following day, cells are washed three times with serum-free medium and

serum starved at 37for 30 min.

3 Coverslips are then inverted onto 50 mL drops of serum free medium

con-taining fluorescently labeled transferrin and incubated for 10 min at 37

(pulse)

4 Coverslips are washed three times with cold complete medium to remove any

excess fluorescently labeled transferrin

5 Coverslips are then inverted onto 50 mL drops of pre-warmed complete

medium containing fluorescently labeled Alexa488-dextran (200 mg/mL)

and Alexa594-LDL (5 mg/mL) and incubated for 37(chase).

6 Coverslips are washed three times with cold PBS at chosen time-points, and

fixed with 4% paraformaldehyde for 30 min

7 Individual cells are imaged at 60 magnification using a DeltaVision

deconvolution microscope Fluorescent transferrin, dextran and LDL are

captured in their respective channels and are deconvolved using Applied

Precisions online microscope software Bright field is also captured for cell

outline

Stage #2 Analysis of Images

1 Cell outlines are manually traced using the bright-field image and added to

ROI manager

2 Images are split into their individual channels for each of the fluorescently

labeled endosomal cargo and are cropped based off the ROI

3 Calculating colocalization of RE constituentsdFluorescent measurements of

colocalization are determined for each marker pair (dextran/LDL,

transferrin/LDL, transferrin/dextran) using the default colocalization tool

within FIJI, Colocalization Threshold, which implements the Manders

overlap coefficient Relative levels of fluorescent colocalization are

compared over multiple time-points to determine the changes in cargo

coincidence throughout the endosomal network

Experiment 2: Rates of Transferrin Trafficking from the Cell Surface

Stage #1 Experimental Setup and Acquisition

1 RAW264.7 macrophages were seeded on MatTek dishes.

2 The following day, cells are washed three times with serum-free medium and

serum starved at 37for 30 min prior to experimentation Excess medium is

then removed from the outer part of the MatTek dish, which is then wiped

dry, leaving a small volume of medium over center coverslip in order to

minimize incubation volumes for labeled transferrin

3 Fluorescently labeled Alexa647-transferrin diluted in medium at a final

concentration of 10 mg/mL is added directly to the cells for 10 min (pulse)

at 37C.

4 Excess transferrin, after the pulse, is washed away rapidly, the MatTek dish is

placed on ice for transport to the imaging chamber, where it is washed with

5 A single field of view would be captured per/dish at 40 magnification toincrease the number of cells that can be imaged simultaneously

6 Fluorescent channels captured included the tagged transferrin, fluorescently

transfected marker (usually eGFP) and bright field to capture cell outline forlater analysis OAI scans were acquired every 20 s for a period of 20 min asthe transferrin exited the cell (chase)

7 Live cell image files are deconvolved as projections using Applied

Precision’s online microscope software

Stage #2 Analysis of Images - Calculating the Half-Life of Transferrin

1 Image analysis is performed in FIJI open access image software.

2 Images undergo a background subtraction, using the default plug-in with a

rolling ball radius of 50 pixels Previous experimentation has suggested thatminimal bleaching occurs with 20 s time-lapses of fluorescently labeledAlexa-dyes

3 For live cells, image sequences are split into three different color channels

(transferrin-cargo, eGFP (e.g., for an over-expressed Rab) and bright field(cell outline)

4 Cell outline images are generated from the first frame of the bright-field

image sequence The assumption is that cells do not move significantlyduring the acquisition so individual segmentation at each time point is notnecessary Cell outlines are individually traced and selected as their ownregion of interest (ROI)

5 ROIs are then overlaid with the fluorescent transferrin channel and individual

fluorescent measurements are taken at each time point To correct ground intensity variation, a background region was selected and the medianintensity in this region was subtracted from the mean cell region reading, foreach slice

back-6 Intensity values are plotted against time, and multiple cells measurements are

grouped for statistical analysis

7 For the determination of the transferrin half-life we need to solve for the exit rate

constant (Kexit) We have modeled the change in amount of transferrin in a cell

as follows: the rate of change in the amount of the transferrin in the cell overtime is proportional to the amount of transferrin in the cell, that is represented

by the equation dXin/dt=KexitXin Therefor; Xin(t)¼ Xin(0)eKexit(t) where

we measure Xin(t) (total Tfn fluorescence at any timepoint (t)), Xin(0) (Tfnfluorescence at time 0) and use Eulers constant (e) to solve for Kexit, usingMatlab or similar software

2.2.3 Experimental considerations

This method can be applied to a variety of experimental questions; however, there

are a number of experimental considerations:

• Cells should be grown to a semi-confluent state to allow for increased cell

numbers from each single acquisition, however, overcrowding can influence the

ability to individually segment cells to single ROIs

• Cells can be treated with a number of pharmacological inhibitors to test their

effects on recycling in the cells, or secretion of cargo via the REs

explored, fluorescently labeled dextran or LDL can be added in conjunction

with transferrin to look at maturation of early endosomes (EEs) to lysosomes,

and or the effects on sorting of cargo as it passes through REs

• Manipulation of host proteins with the use of dominant negative Rab constructs,

or siRNA knockdown of host trafficking Rabs can also be used to explore

potential defects in cell recycling pathways using either fixed RE constituents or

live cell transferrin exit quantification

traf-ficking events of different cargo components throughout the entire endosomal

network The development of this quantification strategy is described elsewhere

in more detail (Lamberton, Condon, Stow, & Hamilton, 2014)

2.2.4 Results

Alexa647-transferrin internalized for 10 min labels an extensive series of punctate or

vesicular structures extending from near the cell surface (labeled for F-actin) to the

Alexa647-transferrin in cells transfected with GFP-Rab11a shows the extensive, but not

com-plete, colocalization of the two markers in RE puncta throughout the cytoplasm

(Figure 2(B)) This shows that internalized transferrin is largely in REs but that,

as a recycling cargo, it does not extend to all parts of the RE network, particularly

the juxtanuclear RE where GFP-Rab11a is more concentrated

Transferrin can be compared to other cargo moving through or intersecting with

the RE network in cells harboring multiple labelled proteins The fluid phase marker,

Alexa488-10,000 MW-dextran is taken up by and labels macropinosomes

Alexa594eLDL is internalized by either macropinocytosis or by receptor-mediated

endocytosis in clathrin-coated vesicles that merge with early/sorting endosomes The

resulting fixed cells show a series of puncta with one, two, or three labeled cargo

Widespread puncta labeled for LDL and/or transferrin (Figure 3(C)) show the

exten-sive reach of the combined early/sorting and RE networks The larger puncta labeled

with dextran indicate the presence of macropinosomes The puncta magnified in the

inset demonstrate how all three cargos intersect in a proportion of endosomes

Performing imaging at different time points of uptake or in pulse-chase regimes,

can be used to chart and quantify the dynamic course of cargo moving through EEs

FIGURE 2 Recycling endosome cargo and trafficking in macrophages.

RAW264.7 macrophages trafficking the well-known recycling endosome cargo Transferrin.(A) Macrophages pulsed with Alexa647-transferrin (pseudo colored green) for 10 min andchased into REs for 20 min were fixed and F-actin labeled with Alexa594 phalloidin (red), thenuclei are labeled with DAPI (blue) The presence of internalized transferrin in REsthroughout the cytoplasm is evident (B) Macrophages transiently transfected with GFP-Rab11a (green) and pulsed with Alexa647-transferrin (red) both label REs, the plasmamembrane is labeled with Alexa350-wheat germ agglutinin (blue) Rab11a and transferrinoverlap in some REs but are not identical, highlighting the dynamic and differentialdistribution of elements within this network (C) Macrophages pulsed with Alexa647-transferrin (pseudo-colored blue) to label REs, Alexa488-dextran labeling macropinosomes(pseudo-colored red) and fluorescently tagged Alexa594-LDL (pseudo-colored green)showing receptor-mediated uptake over 20 min chase Inset shows tri-labeled RE (arrows)containing transferrin, dextran, and LDL (D) Quantification of colocalization betweenfluorescently labeled transferrin (Tfn), dextran (dex), and LDL over time Manders overlapcoefficient is represented as % colocalization (E) Live cell image acquisition of macrophagespulsed with fluorescently labeled Alexa647-transferrin, with frames showing exit oftransferrin over acquisition period (F) Model for exit quantification of transferrin Xin(totalintracellular transferrin), Xout(extracellular transferrin), Kexit(arrow, rate of loss of transferrin).(G) Graphical comparison of experimental data (blue) of transferrin exit over time, andmathematically modeled decay (red) of transferrin exit, with a predicted rate constant of0.0023116, which equates to a transferrin half-life of 300 s (See color plate)

cup and on the phagosome after internalization (arrows) Another internalized IgG-bead is seen in the vicinity with GFP-VAMP3 enrichment on

the phagosome which gradually dissociates from the maturing phagosome (asterisks) (B) Schematic diagram of fluorescence dynamics of

SEP-VAMP3 showing quenching of SEP-SEP-VAMP3 in acidic endosomal compartments SEP-SEP-VAMP3 fluoresces brightly upon exposure to the external

environment (pH 7.4) when recycling endosomes fuse at the plasma membrane (Right panel) RAW264.7 macrophages transiently expressing

SEP-VAMP3 show SEP-VAMP3 on peripheral compartments, on surface ruffles (arrow) and on early phagosomes (box) The phagocytic

membrane around an IgG-opsonized bead shows fluorescence “bursts” of SEP-VAMP3 denoting RE fusion at the phagocytic cup (arrowhead)

More mature phagosomes in the inner part of the cytoplasm are not fluorescing, having lost VAMP3 or having SEP-VAMP3 quenched by the

acidic lumens Scale bar¼ 10 mm (See color plate)

and onto REs and Lysosomes The results depicted inFigure 2(D)record the fates

of transferrin, dextran, and LDL internalized simultaneously in macrophages Cellswere fixed at intervals over 30 min to quantify the fluorescence for each cargo, andcolocalization of pairs of cargo proteins are demonstrated Over time, dextran andLDL, which pass through REs but are largely destined for lysosomes, haveincreasing colocalization with each other Transferrin initially and transiently in-creases in colocalization with LDL, and to a lesser extent with dextran, as thesecargo proteins traverse before being sorted to REs and lysosomes, moving awayfrom the transferrin in the recycling system This demonstrates the concept oftracking cargo moving through the endocytic system and pinpoints key steps thatare required for sorting to REs

Live imaging of internalized fluorescent transferrin is used to chart its uptake fromthe surface as a bolus, its recycling course and exit from the cell over time, which re-sults in a progressive diminution of total intracellular transferrin (Figure 2(E)) Theloss of transferrin fluorescence can be modeled (Figure 2(F) and (G)) Actual fluo-rescence measurements (blue line, Figure 2(G)) show the results concur with themodel of predicted exit (red line,Figure 2(G)) of transferrin from REs By our mea-surements, transferrin typically exits macrophages with a half-life of approximately250e300 s In conclusion, tracking cargo is a powerful strategy for studying thenature and functions of REs in this highly dynamic network in macrophages

2.3.1 Background and objective

REs connect with the cell surface and with membrane protrusions and invaginationsforming directly from the plasma membrane The process of phagocytosis is used bymacrophages for internalizing and destroying pathogens, dead cells, or particles(Botelho & Grinstein, 2011) The plasma membrane and a variety of endosomes,including REs, contribute membrane for the extension of phagocytic cups andthen endosomes participate in the two-way exchange of membrane during matura-tion of the phagosome (Huynh et al., 2007) Different membrane domains labeledwith Rab11a or Rab8a associate with forming phagosomes or their precursor rufflesfor signaling from Toll-like receptors to activate macrophages during pathogenresponses (Husebye et al., 2010; Luo et al., 2014) Thus fusion of REs with earlystage phagosomes is critical for multiple facets of the innate immune response.The membrane fusion protein, VAMP3da member of the SNARE proteinfamilydis a well-established resident of REs in macrophages (Stow et al., 2006).VAMP3 coordinates fusion of the REs with membranes at the base of nascent phago-cytic cups in macrophage (Bajno et al., 2000; Murray et al., 2005) TransmembraneVAMP3 can be labeled with GFP attached to its short lumenal domain (Figure 3(A))and this allows localization of GFP-VAMP3 throughout the cell Alternatively,VAMP3 can be tagged with a pH-sensitive variant of GFPdsuper-ecliptic pHluorin(SEP)dwhich fluoresces at neutral pH but not in the acidic lumens of REs or otherendosomes (Sankaranarayanan, De Angelis, Rothman, & Ryan, 2000) Using this

construct, the actual fusion of REs with phagosomes will be registered by the sudden

fluorescing of the SEP-VAMP3 when the probe is exposed to the neutral pH of the

external medium or in the very early stage phagosome The goal of this experiment

was to image the behavior of GFP-VAMP3-labeled REs in the vicinity of nascent

phagosomes and second to use SEP-VAMP3 to image the fusion of REs with early

stage phagosomes

2.3.2 Flow of the experiment

1 Transfect RAW264.7 macrophages plated on MatTek dishes with GFP- or

SEP-VAMP3 Maintain cells at 37C for 24 h Cells should be 80e90% confluent

upon imaging

2 Activate cells with 100 ng/mL LPS for 30 min before imaging to enhance the

phagocytic capacity of the macrophages

3 Prepare twice the concentration of IgG beads in RPMI 1640 medium (4 mL of

10% IgG bead suspension in 1 mL of medium)

4 Equilibrate MatTek dishes and IgG bead containing culture medium in heating

chamber for 10 min

5 Withdraw 1 mL of culture medium from the MaTtek dish.

6 Choose an optimally spaced and spread out cell expressing SEP-VAMP3 The

cells should not be very bright as SEP-VAMP3 fluorescence is significantly

quenched at 5.5e6.5 low intracellular pH

7 Depending on the fluorescence intensity, the emission power and excitation

time can be ideally reduced as SEP-VAMP3 fluoresces brightly at the plasma

membrane upon fusion of SEP-VAMP3 positive REs

8 Set desired top and bottom Z-slice at 0.2e0.4 mm/step for OAI scan of 3e4 mm

slice

9 Set up image capturing parameters at one frame/s for 10 min.

10 Add 1 mL of IgG bead culture medium.

11 Allow IgG beads to settle down onto cells This usually takes 2e4 min.

The IgG beads can be observed using DIC to track their movement on cells

12 Upon cell contact, imaging capturing can be initiated.

2.3.3 Experimental considerations

The focal depth of phagocytosis might change during the process as the cell draws

the IgG bead into the cell (from top to bottom) Thus manual operation of the

fine-tuning knob on the microscope can be used to accommodate this Latex beads

usu-ally exhibit an optical fluorescence aberration (spherical aberration) around the

circumference of the bead, however the bright VAMP3-SEP fluorescence “bursts”

at the cell surface are obvious even superimposed on this optical effect To avoid

the optical aberration obtained with beads, phagocytosis experiments can be done

using IgG-opsonized sheep red blood cells which do not exhibit this aberration

These image-based methods can be used for rapid, quantitative screening of

phagocytosis in macrophages under drug-treatment regimes or to assess loss of tein function (Yeo, Wall, Stow, & Hamilton, 2013).

pro-2.3.4 Results

Expression of GFP-VAMP3 can be used to localize REs and to track their movement

in live cells (Figure 3(A)) In macrophages, the distribution of GFP-VAMP3 mirrorsthe trafficking cycle of VAMP3 GFP-VAMP3 resides on the perinuclear and punc-tate REs through the cell from where it functions to fuse REs with the cell surface orphagosomes GFP-VAMP3 appears transiently on these destination membranesbefore it is reinternalized and returned to the REs for subsequent rounds of fusion.Accordingly, GFP-VAMP3 is concentrated (transiently) on surface ruffles wheremembrane fusion is prominent (Figure 3(A)) When cells begin to phagocytose sur-face-attached opsonized beads, GFP-VAMP3 appears early on at the base of the beadand GFP-VAMP3-labeled membranes can be seen gradually encircling the beads(Figure 3(A), arrow) During internalization of the phagocytic cup, there is signifi-cant remodeling of the membranes and GFP-VAMP3 is on ruffles, tubules and ve-sicular structures as membrane is exchanged between the phagosome andendosome and the cell surface At later stages of phagocytosis, GFP-VAMP3 islost from the mature phagosome (Figure 3(A), asterisk)

VAMP3 with a C-terminal SEP tag can be used to track the exocytosis of REs,fluorescing as the REs fuse with the plasma membrane or with the still neutral earlyphagosomal membranes before phagosomes mature and acquire an acidic lumen(Figure 3(B)) Newly synthesized SEP-VAMP3 fluoresces moderately in the nearneutral pH of the Golgi complex as it makes its way out to the REs, but then its fluo-rescence is suppressed in the slightly acidic environs of the trans-Golgi network,post-Golgi vesicles, and REs Some puncta near the cell surface fluoresce perhaps

as surface-connected REs or newly internalized endosomes prior to acidification.Strongly fluorescent SEP-VAMP3 is present on surface ruffles (Figure 3(B), arrow)

as the probe is exposed to the neutral pH of the external milieu During phagocytosis(Figure 3(B)), SEP-VAMP3 fluoresces brightly on early stage phagosomes eitherprior to closure, or immediately after closure However, as the phagosome maturesand becomes acidified, SEP-VAMP3 no longer fluoresces The point of fusion of aSEP-VAMP3-labeled RE with the phagosome is seen as a bright burst or spot of fluo-rescence (Figure 3(B), arrowhead) This approach provides direct evidence that REsfuse with early stage phagosomes Thus, this construct is especially useful forobserving fusion of REs at the cell surface in the context of exocytosis and secretion

SUMMARY

In this chapter, we have described several approaches for the imaged-based analysis

of the RE network in macrophages The detailed methodology presented will aidreplication of the RE labeling and analysis approaches we have optimized for use

in macrophages as an example of a highly differentiated, “professional trafficking”

cell type The approaches and results can be extrapolated to other cell types and

indeed it would be interesting to do so as a means of comparing the relative

contri-butions of REs to cell type specific functions Live cell imaging using deconvolution,

confocal, or spinning disc microscopes provides a wealth of dynamic, structural, and

functional information about the REs and RE-associated molecules in macrophages

We have provided snapshots of three approaches for labeling and imaging REs, each

of which can be extended to perform complex analyses or modified for rapid

screening purposes Comparing sets of Rabs by coexpression and colocalization is

a key means for revealing the diverse subcompartments of REs and cells will tolerate

for a short time, transient coexpression of up to five or six labeled Rabs to extend this

type of analysis An endless array of labeled cargo can be employed for tracking

exocytic and endocytic routes through the cell, many of which involve REs We

have provided one set of algorithms for quantifying cargo movement in live cells,

focused here on the transit and exit of a bolus of transferrin In addition to Rabs,

SNAREs are also useful experimentally as organelle markers and here we

demon-strate the use of labelled VAMP3 to show fusion of REs with phagosomes Moreover

the use of the pH-sensitive label is ideal for imaging exocytosis and stage-specific

phagosome delivery The use and extension of all these imaging approaches will

continue to be a mainstay for exploring REs as a multifunctional cell organelle

and for understanding complex RE roles in innate immune responses

REFERENCES

Bajno, L., Peng, X R., Schreiber, A D., Moore, H P., Trimble, W S., & Grinstein, S (2000)

Focal exocytosis of VAMP3-containing vesicles at sites of phagosome formation The

Journal of Cell Biology, 149, 697e706

Beaumont, K A., Hamilton, N A., Moores, M T., Brown, D L., Ohbayashi, N.,

Cairncross, O., et al (2011) The recycling endosome protein Rab17 regulates

melano-cytic filopodia formation and melanosome trafficking Traffic, 12, 627e643

Botelho, R J., & Grinstein, S (2011) Phagocytosis Current Biology, 21, R533eR538

Chavrier, P., Parton, R G., Hauri, H P., Simons, K., & Zerial, M (1990) Localization of low

molecular weight GTP binding proteins to exocytic and endocytic compartments Cell, 62,

317e329

Das, A., & Guo, W (2011) Rabs and the exocyst in ciliogenesis, tubulogenesis and beyond

Trends in Cell Biology, 21, 383e386

Epelman, S., Lavine, K J., & Randolph, G J (2014) Origin and functions of tissue

macrophages Immunity, 41, 21e35

Hsu, V W., & Prekeris, R (2010) Transport at the recycling endosome Current Opinion in

Cell Biology, 22, 528e534

Husebye, H., Aune, M H., Stenvik, J., Samstad, E., Skjeldal, F., Halaas, O., et al (2010) The

Rab11a GTPase controls Toll-like receptor 4-induced activation of interferon regulatory

factor-3 on phagosomes Immunity, 33, 583e596

Hutagalung, A H., & Novick, P J (2011) Role of Rab GTPases in membrane traffic and cell

physiology Physiological Reviews, 91, 119e149

Huynh, K K., Kay, J G., Stow, J L., & Grinstein, S (2007) Fusion, fission, and secretionduring phagocytosis Physiology (Bethesda), 22, 366e372.

Lacy, P., & Stow, J L (2011) Cytokine release from innate immune cells: association withdiverse membrane trafficking pathways Blood, 118, 9e18

Lamberton, T O., Condon, N D., Stow, J L., & Hamilton, N A (2014) On linear models andparameter identifiability in experimental biological systems Journal of TheoreticalBiology, 358, 102e121

Luo, L., Wall, A A., Yeo, J C., Condon, N D., Norwood, S J., Schoenwaelder, S., et al.(2014) Rab8a interacts directly with PI3Kg to modulate TLR4-driven PI3K and mTORsignalling Nature Communication, 15(5), 4407.http://dx.doi.org/10.1038/ncomms5407.Manderson, A P., Kay, J G., Hammond, L A., Brown, D L., & Stow, J L (2007) Subcom-partments of the macrophage recycling endosome direct the differential secretion of IL-6and TNF{alpha} The Journal of Cell Biology, 178, 57e69

Marchese, A (2014) Endocytic trafficking of chemokine receptors Current Opinion in CellBiology, 27, 72e77

Maxfield, F R., & McGraw, T E (2004) Endocytic recycling Nature Reviews MolecularCell Biology, 5, 121e132

Murray, R Z., Kay, J G., Sangermani, D G., & Stow, J L (2005) A role for the phagosome

in cytokine secretion Science, 310, 1492e1495

Nair-Gupta, P., Baccarini, A., Tung, N., Seyffer, F., Florey, O., Huang, Y., et al (2014) TLRsignals induce phagosomal MHC-I delivery from the endosomal recycling compartment

to allow cross-presentation Cell, 158, 506e512

Sankaranarayanan, S., De Angelis, D., Rothman, J E., & Ryan, T A (2000) The use ofpHluorins for optical measurements of presynaptic activity Biophysics Journal, 79,2199e2208

Stenmark, H (2009) Rab GTPases as coordinators of vesicle traffic Nature Reviews ular Cell Biology, 10, 513e525

Molec-Stow, J L., Manderson, A P., & Murray, R Z (2006) SNAREing immunity: the role ofSNAREs in the immune system Nature Reviews Immunology, 6, 919e929

Stow, J L., & Murray, R Z (2013) Intracellular trafficking and secretion of inflammatorycytokines Cytokine Growth Factor Reviews, 24, 227e239

Su¨dhof, T C., & Rothman, J E (2009) Membrane fusion: Grappling with SNARE and SMproteins Science, 323, 474e477

Wickner, W (2010) Membrane fusion: five lipids, four SNAREs, three chaperones, twonucleotides, and a Rab, all dancing in a ring on yeast vacuoles Annual Reviews Celland Developmental Biology, 26, 115e136

Yeo, J C., Wall, A A., Stow, J L., & Hamilton, N A (2013) High-throughput quantification

of early stages of phagocytosis Biotechniques, 55, 115e124

Analyzing the functions of

Rab11-effector proteins

Rytis Prekeris

Department of Cell and Developmental Biology, University of Colorado Anschutz Medical Campus,

Aurora, CO, USA E-mail: rytis.prekeris@ucdenver.edu

2 Time-Lapse Analysis of Endosome and Actin Dynamics during Cytokinesis 27

2.1 Analyzing Rab11/FIP3-Endosome Transport during Cytokinesis 27

2.1.1 Procedure 28

2.1.2 Data interpretation 29

2.2 Analyzing Actin or Microtubule Dynamics during Cytokinesis 29

2.3 Cytokinesis and Apical Lumen Formation in 3D Cultures 30

Recycling endosomes recently have emerged as major regulators of cytokinesis and

abscission steps of cell division Rab11-endosomes in particular were shown to transport

proteins to the mitotic ingression furrow and play a key role in establishing the abscission

Methods in Cell Biology, Volume 130, ISSN 0091-679X, http://dx.doi.org/10.1016/bs.mcb.2015.05.001

site Rab11 GTPase functions by binding and activating various effector proteins, such asRab11 family interacting proteins (FIPs) FIPs appear to be at the core of many Rab11functions, with FIP3 playing a role in targeting of the Rab11-endosomes during mitosis.Here we summarize the newest finding regarding the roles and regulation of FIP3 andRab11 complex, as well as describe the methods developed to analyze membrane andcytoskeleton dynamics during abscission step of cytokinesis.

INTRODUCTION

The last step of cell division is a physical separation of two daughter cells via aprocess known as cytokinesis (Barr & Gruneberg, 2007; Pollard, 2010) Afterreplication of the genetic material, the mother cell divides by the formation ofthe cleavage furrow that constricts cytoplasm leaving two daughter cells connected

by a thin intracellular bridge (ICB) The resolution of this bridge (abscission)results in separation of two daughter cells Although the mechanisms that governabscission are not fully understood, recent evidence suggests that actin cytoskel-eton, endosomes, and the ESCRT-III protein complex play a critical role in thisprocess

ESCRT complexes (complexes 0, I, II, and III) were originally described as ulators of multivesicular body formation (Babst, Katzmann et al., 2002) Since thenseveral ESCRT proteins, namely Tsg101, Alix, and ESCRT-III complex proteins,were demonstrated to be required for cytokinesis (Carlton, Agromayor et al.,2008; Carlton & Martin-Serrano, 2007) The model of ESCRT recruitment to theICB is as follows: Alix and/or Tsg101 are recruited to the midbody by binding tothe midbody protein CEP55 These components then recruit various ESCRT-IIIcomplex members to the midbody The ESCRT-III complex has the ability to

et al., 2011; Guizetti, Schermelleh et al., 2011) How ESCRT-III complex proteinsmove from the midbody to the abscission site remains unclear, but we have shownthat localized actin depolymerization and narrowing of the ICB (secondary ingres-sion) are required to establish the abscission site and recruit the ESCRT-III complex(Figure 1)

Recycling endosomes (RE) have emerged as important players in mediatingabscission (Fielding, Schonteich et al., 2005; Schiel, Childs et al., 2013; Wilson,Fielding et al., 2004) Several reports demonstrated that pronounced changes occur

in endocytic recycling during mitosis, and that these changes are required for cessful completion of cytokinesis Originally, it was proposed that REs initiateabscission by fusing with each other and the plasma membrane, thus building a sepa-rating membrane in a manner similar to a formation of the phragmoplasts in plantcells However, recent data from our laboratory (Schiel, Park et al., 2011; Schiel,Simon et al., 2012) have shown that fusion of REs mediates the formation of a “sec-ondary ingression,” thus initiating ESCRT-III recruitment to the abscission site(Figure 1)

suc-membrane Among other factors, these organelles deliver OCRL and p50RhoGAP leading tothe localized disassembly of actin cytoskeleton and severing of central spindle microtubules.Actin depolymerization induces formation of the secondary ingression and ESCRT-IIItranslocation from the midbody to the abscission site Delivery of the ESCRT-III to thesecondary ingression leads to a final scission event and separation of daughter cells.

Rab11, a small GTPase that functions in RE-mediated trafficking of plasmamembrane receptors, has recently emerged as a key regulator of RE transport tothe ICB during abscission (Fielding et al., 2005; Wilson et al., 2004) All RabGTPases function by binding and recruiting various effector proteins While severalRab11-effector proteins have been identified, Rab11 regulates RE delivery to theICB predominantly via binding to its FIP3-effector protein (Fielding et al., 2005;Wilson et al., 2004) The FIP3/Rab11 complex accumulates at the ICB duringmitosis, and depletion of FIP3 by siRNA arrests cells in late telophase, while having

no effect on anaphase/early telophase (Simon, Schonteich et al., 2008; Wilson et al.,

2004) Interestingly, recent data show that FIP3-endosomes deliver p50RhoGAP(known RhoA GAP) to the ingression furrow during late telophase Upon delivery

of p50RhoGAP to the ICB, it mediates nested depolymerization of the actin skeleton, leading to the formation of the “secondary ingression” and abscission(Schiel et al., 2012) (Figure 1)

cyto-The past few years have seen a dramatic increase in our understanding of sion It is clear that highly dynamic regulation of endosomes, cytoskeleton, andESCRT complexes is the key to the successful completion of cytokinesis (Schiel

abscis-et al., 2013) Here, we describe a set of novel techniques/approaches that will allowthe further dissection of the machinery governing cytokinesis and abscission

CYTOKINESIS

The studies from many research groups led to development of several abscissionmodels These studies are based on the use of various techniques The temporary na-ture of the abscission event makes it difficult to study Furthermore, the abscissionsite is very small and difficult to image using standard light microscopy techniques.Finally, there are currently no good markers to clearly identify the abscission site.Consequently, most of the techniques used to detect the abscission defects usuallymeasure the failure of cells to separate rather than the actual abscission event.Unfortunately, the separation of daughter cells is a very complex event that can

be affected by many factors Accordingly, each of the techniques (described bellow)has its shortcomings and a combination of multiple approaches should be used to testthe involvement of the candidate protein in the abscission event

One of the most commonly used and easiest techniques for investigating proteinsthat regulate cytokinesis is to measure the multinucleation of cells The idea behindthis approach is that the inhibition of abscission results in the regression of the cleav-age furrow, thus leading to bi- or multinucleated cells This technique has beencommonly used to detect defects in the formation and ingression of the cleavagefurrow One drawback of using this approach to analyze the abscission event is

the failure of cleavage furrow to regress in cells arrested at late telophase Instead,

these cells stay connected with the ICB, often long enough to start a second round of

division Therefore, to better analyze the effect on cell abscission, cells are usually

scored for the number of bi- and multinucleated cells along with the number of cells

connected by ICBs

1.1.1 Procedure

1 Coat 22 mm glass coverslips with collagen (placed in 6-well dish) Make sure

to clean the glass coverslips with 3% acetic acid, followed by wash with water,

beforehand to ensure an even coating of collagen Alternatively, poly-L-lysine

can also be used as a substrate

2 Under a cell culture hood, plate HeLa cells on collagen-coated glass coverslips.

Make sure that cells are no more than 30% confluent Plating cells too densely

will affect their division rates, thus directly affecting the levels of

multi-nucleation It is paramount to always plate equal number of cells, since the cell

density will directly affect cell division rates

3 Incubate cells in serum-supplemented media for 24 h.

4 Rinse cells with phosphate-buffered saline (PBS) and fix for 15 min with 4%

paraformaldehyde (in PBS)

5 Permeabilize cells with 2 mL of blocking solution (PBS, 2% fetal bovine

serum, 1% albumin, 2% saponin) by incubating for 20 min

6 Incubate cells with primary anti-acetylated tubulin antibody (SigmaeAldrich,

cat#T7451; dilution 1:100) for 1 h by overlaying 100mL of blocking solution

with primary antibody Make sure to place a small piece of parafilm on top of

the coverslip and wrap 6-well dish in moist paper towel to minimize

evaporation

7 Wash cells three times (5 min each) with 2 mL of blocking solution Overlay

cells with 100mL of blocking solution with secondary antibody, DAPI, and

rhodamine-phalloidin (Invitrogen, cat#R415; dilution 1:50) We typically use

anti-mouse IgG antibody conjugated to Alexa488 Incubate for 30 min

8 Wash cells three times (5 min each) with 2 mL of blocking solution.

9 Mount cells in Vectashield (Vector Laboratories, cat#H-1000) Seal edges with

nail polish

10 Image cells using florescent micropscopy Rhodamine-phalloidin will allow

you to visualize the edges of the cells to count the number of nuclei within

each cell Similarly, anti-acetylated tubulin antibodies will stain the central

spindle, allowing visualization of the ICBs

1.1.2 Data interpretation

Typically, untreated HeLa cells have around 2e5% of multinucleated cells Similarly,

3e6% of cells are still connected with the ICB and therefore in the late telophase

stage Knockdown of known key regulators of abscission, such as FIP3 or CHMP4B

(ESCRT-III subunit) usually results in about 20% multinucleation and about 25%

cells that remain connected with ICBs (Carlton et al., 2008; Schiel et al., 2013)

This is presumably due to the fact that some cells that are arrested in late telophasewill eventually undergo apoptosis Indeed, FIP3 knockdown leads to a significantdecrease in cell survival (Schiel et al., 2012) Interestingly, a large portion of thetelophase-arrested cells will resolve their ICBs using traction-dependent cytokinesis,generating forceful breakage of the extended ICBs and the eventual separation ofdaughter cells Thus, if the candidate protein is involved in abscission, one shouldnot expect to see more than 20e25% multinucleated cells.

Due to the redundancy in cytokinesis machinery knockdown of any single proteininvolved in mediating abscission may not completely block cell division A moredirect method of evaluating the involvement of candidate proteins in mediatingthe abscission event is to measure the time required for daughter cells to completecytokinesis Indeed, many well-established abscission regulators dramaticallyincrease division time while having a very moderate effect on multinucleation(Dambournet, Machicoane et al., 2011; Schiel et al., 2012)

1.2.1 Procedure

1 Plate Hela cells on collagen-coated glass bottom Grid-50 dishes (Ibidi,

cat#81148) Make sure that cells are plated at 20e30% confluency since cellswill be imaged for 24 h, during which time the majority of cells will divide atleast once Plating cells too densely will make it difficult to visualize the timing

of abscission

2 Let cells attach for 3e4 h in serum-supplemented media Replace media toremove floating cells

3 Set up cells for imaging on an inverted microscope equipped with an X-Y-Z

motorized stage and an environmental control system

4 Image cells with a 20X objective Pick at least five to six imaging fields (with

10e15 cells each) and image by field-contrast at 30 min time-lapses for 24 h

5 During cell division the ICB and midbody are quite dynamic and often can leave

the field of focus To ensure visualization of each abscission event at every timepoint a mini-Z-stack should be imaged Typically taking 10 images separated by

1mm Z-step is sufficient to ensure that ICB and midbody can be evaluated atevery time point

1.2.2 Data interpretation

Once imaging is complete, the time required for cytokinesis is measured for everycell in each field Typically, the metaphase is considered to be time point 0, whilethe resolution of the ICB is marked as the last step in daughter cell separation.The time required for cells to complete mitosis is quite variable and can rangefrom 60 min to 3e4 h As a result, a large sample of cells is needed (80e100 cellsfor every condition) to derive a meaningful and statistically sound data about thetiming of cytokinesis

1.3 ANALYZING THE ESTABLISHMENT OF THE ABSCISSION SITE

Measuring the formation of abscission site is the most direct way of testing the effect

of candidate proteins in regulating abscission Endosomes initiate the abscission site

by regulating localized depolymerization of actin cytoskeleton by either delivering

(Rab35-endosomes) (Dambournet et al., 2011; Schiel et al., 2012) Actin depolymerization

then leads to the formation of the secondary ingression, leading to the formation of

the abscission site (Schiel et al., 2011) The abscission site can be identified by

several methods First, central spindle microtubules are heavily acetylated and are

clearly identifiable following staining with anti-acetylated tubulin antibody Since

abscission involves localized spastin-dependent cutting of microtubules (Connell,

Lindon et al., 2009) the abscission site can be identified as a gap in the central

spin-dle (Elia et al., 2011; Schiel et al., 2011) Note that the central spindle stained with

anti-acetylated tubulin antibodies always has a gap in the middle that is caused by

limited access of antibodies to a dense midbody structure The abscission site

usu-ally forms on one or both sides of this midbody-gap and can be identified as

second-ary gaps in anti-acetylated tubulin staining (Figure 2) Alternatively, the ESCRT-III

FIGURE 2 Models of asymmetric and symmetric abscission.

In many dividing cells abscission sites are established bilaterally on both sides of the

midbody That leads to a shedding of the midbody to the extracellular space (left panel)

However, abscission site can also be formed only on one side of the midbody, leading to

asymmetric abscission (right panel) This type of asymmetric abscission results in midbody

inheritance by one of the daughter cells

complex can also be used as an abscission site marker Typically, CHMP4B-GFP isused in these types of studies, since good anti-CHMP4B antibodies are not available.Along with other ESCRT complex members, CHMP4B-GFP is first recruited to themidbody (Figure 1) (Elia et al., 2011; Schiel et al., 2012) After formation of thesecondary ingression, part of CHMP4B-GFP translocates to the abscission siteand can be identified as a puncta on one or both sides of the midbody If cells arecostained with anti-acetylated tubulin antibodies, these CHMP4B puncta will bepresent in the secondary gaps within central spindle microtubules (Elia et al.,2011; Schiel et al., 2012).

1.3.1 Procedure

1 Under a cell culture hood, plate HeLa cells on collagen or polycoated glass

coverslips Make sure that cells are no more than 30% confluent Plating cellstoo densely will make it difficult to visualize central spindle and the midbody

2 Incubate cells in serum-supplemented media for 24 h.

3 If using CHMP4B-GFP as an abscission marker, transfect cells and incubate for

another 28 h

4 Rinse cells with phosphate-buffered saline (PBS) and fix for 15 min with 4%

paraformaldehyde (in PBS)

5 Permeabilize cells with 2 mL of blocking solution (PBS, 2% fetal bovine

serum, 1% albumin, 2% saponin) by incubating for 20 min

6 Incubate cells with primary anti-acetylated tubulin antibody (SigmaeAldrich,cat#T7451; dilution 1:100) for 1 h by overlaying 100mL of blocking solutionwith primary antibody Make sure to place a small peace of parafilm on top ofthe coverslip and wrap 6-well dish in a moist paper towel to minimizeevaporation

7 Wash cells three times (5 min each) with 2 mL of blocking solution Overlay

cells with 100mL of blocking solution with secondary antibody and DAPI Wetypically use anti-mouse IgG antibody conjugated to Alexa594 to allow co-imaging of acetylated tubulin with CHMP4B-GFP Incubate for 30 min

8 Wash cells three times (5 min each) with 2 mL of blocking solution.

9 Mount cells in Vectashield (Vector Laboratories, cat#H-1000) Seal edges with

nail polish

10 Image cells by fluorescent microscopy Anti-acetylated tubulin antibodies will

stain the central spindle If cells were transfected with CHMP4B-GFP, it canalso be used to identify abscission sites

1.3.2 Data interpretation

The abscission rates can be evaluated by counting the percentage of telophase cellsthat have clearly established an abscission site, as determined by the presence of agap in the central spindle Care must be taken not to count a midbody-associatedgap Similarly, abscission sites can be counted by determining the percentage ofthe cells that contain CHMP4B-GFP puncta within the ICB, but outside the mid-body In addition to counting cells that have an established abscission site, one

can also analyze the effect of experimental treatment on the rates of asymmetric

abscission In about 40% of HeLa cells abscission usually occurs on both sides of

the midbody leading to the release of the midbody into the media (Figure 2)

Inter-estingly, in a majority of the cell divisions the abscission site is only established on

one side of the midbody (Kuo, Chen et al., 2011) As a result of this asymmetric

abscission, one daughter cell will inherit the postmitotic midbody (Figure 2)

Impor-tantly, midbody inheritance has been proposed as a mechanism for regulating cell

fate and differentiation in daughter cells (Chen, Ettinger et al., 2012; Ettinger,

Wilsch-Brauninger et al., 2011; Kuo et al., 2011) Thus, it will be important to

iden-tify the factors that mediate the establishment of these asymmetric abscissions

One of the potential drawbacks of using CHMP4B-GFP overexpression as an

abscission site marker is the fact that high levels of CHMP-GFP can inhibit

abscis-sion presumably by interfering with ESCRT-III function Thus, care must be taken to

analyze only cells expressing low to moderate levels of CHMP4B-GFP

DYNAMICS DURING CYTOKINESIS

Microscopy analysis of fixed cells remains a powerful tool for the initial

character-ization of candidate proteins that regulate abscission However, static image analysis

will not uncover the properties and regulation of dynamic changes in action

cyto-skeleton and membrane dynamics during cell division Thus, time-lapse analysis

of live dividing cells is needed to investigate rapid and localized changes in

cytoskel-eton and endosomes during abscission The approaches listed below will allow

re-searchers to investigate the changes in Rab11/FIP3-endosomes (A) or actin and

microtubules (B)

CYTOKINESIS

Rab11/FIP3-endosomes have recently emerged as key mediators of abscission

(Wilson et al., 2004; Schiel et al., 2012) Importantly, the dynamic movement and

localization of Rab11/FIP3-endosomes change dramatically as cells proceed from

metaphase to telophase (Figure 3) During metaphase Rab11/FIP3-endosomes

accu-mulate around centrosomes, where they stay until telophase (Figure 3) (Schiel et al.,

2011) After cells form the midbody and start progressing to late telophase, Rab11/

FIP3-endosomes move along the central spindle microtubules to the cleavage furrow

(Figure 3) (Schiel et al., 2011; Simon et al., 2008) This leads to the translocation of

all Rab11/FIP3-endosomes from centrosomes to the endocytic pools situated on each

side of the midbody (Figure 3) The translocation of the Rab11/FIP3-endosomes is

usually nonsymmetric, with endosomes from one centrosome arriving at the

mid-body slightly ahead of the endosomes from the other side (Schiel et al., 2011) Since

Rab11/FIP3-endosomes initiate the formation of the secondary ingression and theestablishment of the abscission site, potentially this asynchronous arrival at thecleavage furrow determines asymmetric abscission and midbody inheritance byone daughter cell Despite the potential impact on cell fate and differentiation, themachinery mediating asynchronous translocation of Rab11/FIP3-endosomes to thecleavage furrow remains essentially unknown Thus, time-lapse analysis of FIP3-GFP dynamics during cell division is perfectly suited to decipher the mechanismsgoverning Rab11/FIP3-endosome dynamics and midbody inheritance.

2.1.1 Procedure

1 Transfect HeLa cells with FIP3-GFP and plate cells on collagen-coated glass

bottom Grid-50 dishes (Ibidi, cat#81148) Make sure that cells are plated at 30e50% confluency to allow clear visualization of FIP3-GFP dynamics Platingcells too dense will result in the formation of short intercellular bridges, makingvisualization more difficult

2 Set up cells for imaging on an inverted microscope equipped with an

environ-mental control system Cells need to be imaged at 37C to ensure properendosome dynamics and cytokinesis Imaging cells at room temperature oftenleads to either mitotic arrest or dramatically increase the time required forabscission

3 Image cells with a 63X oil-objective Pick cells at metaphase and image it every

10e15 min Since during metaphase cells round up, small Z-stack with 500 nmZ-step About 30 stack images will be needed to image the entire cell

4 Imaging cells every 10e15 min will uncover the steady-state changes in GFP localization but will not allow tracking of individual cells To visualize andanalyze the speed and dynamics of individual FIP3-GFP-endosomes 300 mstime-lapses will be needed Unfortunately, imaging dividing metaphase cellsusing such short time-lapses will lead to phototoxicity and can dramaticallyaffect the cell’s ability to divide Therefore, to analyze the dynamics of

FIP3-FIGURE 3 Rab11/FIP3-endosome dynamics during cytokinesis.

During metaphase and anaphase Rab11/FIP3-endosomes associate with centrosomes Asdividing cell progresses to telophase and midbody is formed, Rab11/FIP3-endosomes starttranslocated along central spindle microtubules to the close proximity of the midbody By latetelophase almost all Rab11/FIP3-endosomes are localized at and around the midbody Thistranslocation precedes and is required for the final abscission event

individual FIP3-endosomes we usually pick cells at early telophase that contain

a formed midbody which substantially reduces the number of time points

needed to visualize FIP3-GFP dynamics through the abscission event

Gener-ally, no more than 50e80 time points should be taken for each cell

Further-more, picking cells at telophase results in much flatter cells compared to

metaphase cells, thus eliminating the need to do z-stacks and further reducing

imaging-induced photodamage

2.1.2 Data interpretation

Time-lapse analysis is a powerful tool for investigating many different aspects of

changes in the steady-state Rab11/FIP3-endosome distribution at different mitotic

stages can be captured Using 300 ms time-lapse windows will allow for

measure-ment of the directionality, processivity, and speed of individual

Rab11/FIP3-endosomes

CYTOKINESIS

In addition to changes in Rab11/FIP3-endosome dynamics, the actin and

microtu-bule cytoskeleton also undergo dramatic and very dynamic changes As described

earlier, during telophase central spindle microtubules are compacted together to

form a very dense cellular structure known as the midbody The spastin-dependent

localized severing of these central spindle microtubules is a key step in determining

the location and timing of the abscission Although the machinery determining the

site of spastin-dependent microtubule severing remains unclear the location of “cut”

appears to depend on central spindle bending during late telophase (Simon et al.,

2008)

Actin filaments also dynamically change during cytokinesis During early

anaphase actin forms a filamentous actomyosin contractile “ring” at the midzone

of the dividing cell The contraction of this actomyosin ring leads to the formation

and ingression of the cleavage furrow As ingression progresses, the actomyosin ring

gets compacted leading to the dramatic increase in filamentous actin amount at the

plasma membrane of the cleavage furrow However, once the ingression is complete

and the midbody has formed, the acto-myosin network undergoes a very rapid and

localized disassembly, the step that is controlled by endocytic transport and is

required for the abscission (Dambournet et al., 2011; Schiel et al., 2012)

These findings clearly demonstrate the dynamic nature of cytoskeleton and

sup-port the ides that to fully understand the regulation of cytokinesis it is imperative to

perform time-lapse microscopy to analyze filamentous actin and microtubule

dy-namics Microtubules are typically analyzed using GFP or mCherry-tagged tubulin

This type of analysis usually works better in cells stably expressing tagged-tubulin

Similarly, actin dynamics can be analyzed in cells transfected with GFP-actin

How-ever, some published reports have suggested that GFP-tagging can affect filamentous