Available via license: CC BY 3.0
Content may be subject to copyright.
Immunity, Volume 35
Supplemental Information
Lysosomal Trafficking, Antigen Presentation,
and Microbial Killing Are Controlled
by the Arf-like GTPase Arl8b
Salil Garg, Mahak Sharma, Cindy Ung, Amit Tuli, Duarte C. Barral, David L. Hava, Natacha
Veerapen, Gurdyal S. Besra, Nir Hacohen, and Michael B. Brenner
Figure S1, related to Figure 1: shRNAs targeting pro-Saposin and Arl8b
reduced CD1d antigen presentation
A. Sample library screening plate containing shRNAs targeting pro-Saposin and other
molecules
B: Multiple distinct hairpins silenced Arl8b and reduced CD1d antigen presentation
Presentation of Gal-GalCer by U937 to NKT Cells
0.01
0.1
1
10
100
1000
0
0
1
2
3Control Hairpin #1
Control Hairpin #2
Arl8b Hairpin 347
Arl8b Hairpin 346
Arl8b Hairpin 377
Arl8b Hairpin 407
Arl8b Hairpin 602
Arl8b Hairpin 921
Gal-GalCer ng/mL
IFN ng/mL
Arl8b mRNA levels in U937 stably
transduced with shRNA
Control Hairpin #2
Control Hairpin #1
8b-377
8b-602
8b-407
8b-347
8b-921
8b-346
0.00
0.02
0.04
0.06
Expression relative to GAPDH
Figure S1: shRNAs targeting pro-Saposin and Arl8b reduced CD1d antigen presenation: A. Sample
library screening plate containing shRNAs targeting pro-Saposin: Shown above are averages of three
replicates for each shRNA construct on screening plate 15, which contained hairpins targeting pro-Saposin (red
bars) amongst others. Also highlighted are negative control shRNAs randomly distributed across each screening
plate targeting LacZ, GFP, and RFP (green bars). “% Expected IFN-” was calculated by comparing the IFN-
secreted by NKT-cells for each construct to the best fit line for that plate (as in Figure 1A). “% Expected IFN-
” of 100% indicates a shRNA which had no effect on CD1d antigen presentation. Hairpins targeting pro-
Saposin (red bars) gave strong reductions in CD1d antigen presentation whereas those targeting control genes
not expressed in U937 cells (green bars) did not affect IFN- responses. B. Multiple distinct hairpins silenced
Arl8b and reduced CD1d antigen presentation: We generated a series of shRNA hairpins targeting various
locations along the length of the mature Arl8b transcript. The numerals indicate the position of the first targeted
residue from the 5’ end of the mRNA molecule (see Table 1). U937 cells were stably transduced with lentivirus
expressing each shRNA separately. Left: Cells were plated for antigen presentation with indicated doses of
Gal-GalCer, NKT Cells (BM2a.3) were added, and following overnight incubation response was quantified
by IFN- ELISA. Note all hairpins targeting Arl8b gave significant reductions in dose-dependent antigen
presentation compared to control hairpins. Right: Total RNA preps from stably transduced U937 cells were
analyzed for Arl8b mRNA levels by RT-qPCR. Expression was plotted relative to GAPDH using the
comparative Ct method. All six shRNA targeting Arl8b showed significant reduction in Arl8b mRNA.
Figure S2, related to Figure 2: Arl8b localized to lysosomes and controlled
mCD1d entry
A. Endogenous Arl8b localized to lysosomes
B. Quantification of Arl8b localization to intracellular compartments
Localization of Arl8b to intracellular compartments
LAMP1
EEA1
M-6-PR
CALNEXIN
GM130
0
25
50
75
Colocalization (%) between
Arl8b staining and indicated marker
C. mCD1d was delayed entering lysosomes in Arl8b silenced cells
D. Steady state localization of markers to lysosomes was not impacted by Arl8b
silencing
Figure S2: Arl8b localized to lysosomes and controlled mCD1d entry A. Endogenous Arl8b localizes to
lysosomes: (first row) HeLa cells were stained for LAMP1 (Alexa Fluor 546, red) and with rabbit anti-Arl8
antisera followed by Alexa Fluor-488 (green) conjugated donkey anti-rabbit. The antiserum dominantly
recognized LAMP1+ lysosomes, confirming this as the location of endogenous Arl8b. (second row) In contrast,
the early endosome marker EEA1 showed no colocalization with Arl8b, indicating Arl8b did not localize to
early endosomes. (third row) As in A,B, & C. M-6-PR has a complex intracellular distribution (Golgi, trans-
Golgi, late endosomes, and partially to lysosomes). Arl8b did not show a significant distribution to M-6-PR+
compartments. (fourth row) The ER marker Calnexin showed no colocalization with Arl8b, indicating the
molecule did not localize to this compartment. (fifth row) The Golgi marker GM130 showed no colocalization
with Arl8b, indicating Arl8b did not localize to this compartment. Shown are representative sample images.
Similar results were obtained in RAW macrophage cells. All scale bars=10m. B. Quantification of Arl8b
localization to intracellular compartments: Quantification of colocalization (Metamorph) between Arl8b
antisera staining and the indicated markers for >30 cells for each staining in HeLa cells is shown. The pool of
molecules recognized by Arl8b antisera dominantly colocalized with lysosomal marker LAMP1 and failed to
colocalize significantly with markers of other compartments. C. mCD1d was delayed entering lysosomes in
Arl8b silenced cells: mAb 19G11 was bound to CD1d in Arl8b silenced or control RAW cells at 4°C for 30
min. Cells were then warmed to 37°C for indicated times (0 minutes, 20 minutes, 60 minutes), fixed, and
stained with anti-19G11 secondary (red) and anti-LAMP1 (green). (left panels) Representative images from 60
minute timepoint for Arl8b silenced (bottom row) and control cells (top row). The inset shows strong
accumulation of CD1d in LAMP1+ lysosomes in control cells, and little or no CD1d in Lamp1+ compartments in
Arl8b silenced cells. (right panel) Colocalization for >20 cells for each sample at each timepoint was quantified
as described (methods). D. Steady state localization of markers to lysosomes was not impacted by Arl8b
silencing: (first and second row) HeLa cells were fixed, permeabilized, and stained for saposin C (SAPC,
middle panels) and LAMP1 in control (first row) and Arl8b silenced (second row) cells. Similar to control cells,
Arl8b silenced cells show strong colocalization between SAPC and LAMP1. This suggests that Arl8b silencing
did not result in loss of Saposin molecules from lysosomes. (third and fourth rows) RAW cells were stained for
cathepsin D (middle panels) and LAMP1 in control (third row) and Arl8b silenced (fourth row) RAW cells. In
both Arl8b silenced and control cells, a dominant portion of cathepsin D staining colocalizes with LAMP1.
(fifth and sixth rows) Arl8b silenced (sixth row) and control (fifth row) HeLa cells were transfected with a
Rab9-GFP construct, fixed, and stained for LAMP1 (red, left panels). Rab9 displays a complex distribution,
labeling late endosomes, lysosomes, and trans-Golgi. Note that a portion of Rab9+ vesicles colocalized with
LAMP1+ vesicles in control cells (fifth row). In Arl8b silenced cells (sixth row), a similar portion of Rab9
continued to colocalize with Arl8b. These results suggest localization of a variety of markers to lysosomes
(SAP C, cathepsin D, Rab9) at steady state was unaltered in Arl8b silenced cells.
Figure S3, related to Figure 3: Arl8b amounts controlled the distribution of
lysosomes
A. Arl8b silencing resulted in a mislocalization of lysosomes
B. Arl8a and Arl8b double silencing increased clustering of lysosomes in RAW cells
Figure S3: Arl8b amounts controlled the distribution of lysosomes A. Arl8b silencing resulted in a
mislocalization of lysosomes: Arl8b silenced (middle row) or control HeLa cells (top and bottom rows) were
plated on glass coverslips. (top and middle rows) Cells were fixed, permeabilized, and stained with anti-
LAMP1 to mark lysosomes and anti--tubulin to mark the tubulin cytoskeleton and indicate the boundaries of
the cell. In control cells (top row), LAMP1+ lysosomes distributed throughout the cytoplasm. In contrast,
LAMP1+ lysosomes displayed a marked mislocalization to the perinuclear area of the cell upon Arl8b silencing
(middle row). (bottom row) HeLa cells were transfected with an Arl8b overexpression construct, fixed, and
stained with anti-LAMP1 and anti-Arl8b. An asterisk denotes a transfected cell. In contrast to Arl8b silencing,
Arl8b overexpression resulted in lysosomes redistributing to the periphery of the cell and sometimes forming
clusters there (indicated with arrows). B. Arl8a and Arl8b double silencing increased clustering of
lysosomes in RAW cells: RAW cells were stably transduced with shRNA targeting Arl8b (pLKO.1 TRC1
vector, puromycin resistance) and shRNA targeting Arl8a (pLKO.1 TRC18 vector, blasticidin resistance) giving
Arl8a/8b double knockdown cells (bottom row). Alternatively, RAW cells were stably transduced with two
control sequences expressed from the same vectors (top row). Cells were fixed, permeabilized, and stained with
anti-LAMP1 and anti--tubulin. Arl8 double silenced cells showed a dramatic clustering of LAMP1 when
compared to control cells (compare middle panels in top and bottom row). RAW cells express both Arl8a and
Arl8b and do not exhibit the same degree of lysosome mislocalization as HeLa cells when Arl8b is silenced
alone (compare A middle row to B bottom row). Additionally, Arl8 double silenced RAW cells showed more
dramatic lysosome clustering than Arl8b single silenced RAW cells (compare LAMP1 distribution in B bottom
row above to LAMP1 distribution in Supplemental Figure 2C bottom row and Figure 6A and 6B bottom
rows).
Figure S4, related to Figure 4: Pulldown with GST-Arl8b identified
members of the HOPS complex
Arl8b-interacting proteins
# of unique peptides
Tubulin beta-2B chain
28
Tubulin alpha-1A chain
21
Tubulin beta-4 chain
10
Tubulin beta-6 chain
8
Tubulin beta-5 chain
4
Tubulin beta-3 chain
4
Tubulin alpha-4A chain
3
Tubulin alpha-1B chain
2
Vimentin
7
Annexin A2
5
Capza1 F-actin-capping protein subunit alpha-1
2
ATP5a1 ATP synthase subunit alpha, mitochondrial
17
ATP5b ATP synthase subunit beta, mitochondrial
17
Slc25a4 ADP/ATP translocase 1
11
V-type proton ATPase subunit H
4
Vps16 Isoform 1 of Vacuolar protein sorting-
associated protein 16 homolog
3
Vps18 of Vacuolar protein sorting-associated
protein 18 homolog
2
Prkaa1 protein kinase, AMP-activated, alpha 1 catalytic
subunit
6
Tbc1d9b isoform 1 of TBC1 domain family member 9B
4
Iqgap1 Ras GTPase-activating-like protein IQGAP1
4
Figure S4: Pulldown with GST-Arl8b identified members of the HOPS complex: Lysates from RAW
macrophages were probed with GST-Arl8b or GST. Eluates were run on SDS-PAGE. Bands which appeared
specifically in the GST-Arl8b lane were cut, partially digested, and analyzed by mass spectrometry. The
number of unique peptides identified corresponding to each interaction partner is listed at right. We restricted
our initial analysis to those genes where at least two unique peptides were identified in this manner. Previously,
Arl8b has been shown to bind -tubulin. Encouragingly, we identified a number of tubulin subunits as potential
binding partners of Arl8b. We also identified two members of the VPS-C core of the HOPS complex (see
Figures 4-5).
Figure S5, related to Figure 5: Arl8b and VPS41 promoted the recruitment
of HOPS complex members to lysosomes
A. Arl8b promoted the recruitment of VPS39 to lysosomes
B. Arl8b and VPS41 promoted the recruitment of VPS11 to lysosomes
C. Arl8b and VPS41 promoted the recruitment of VPS16 to lysosomes
Figure S5: Arl8b and VPS41 promoted the recruitment of HOPS complex members to lysosomes A.
Arl8b promoted the recruitment of VPS39 to lysosomes: HeLa cells were transfected with VPS39-HA alone
or VPS39-HA with Arl8b, fixed, and stained with anti-HA (hVPS39) and anti-LAMP1. In control cells,
hVPS39 distributed to the cytosol and did not colocalize with lysosomal marker LAMP1 (top row). In contrast,
Arl8b overexpression resulted in a dramatic recruitment of hVPS39 to lysosomes (bottom row). B. Arl8b and
VPS41 promoted the recruitment of VPS11 to lysosomes: HeLa cells were transfected with VPS11-GFP (top
row), VPS11-GFP + Arl8b (middle row), or VPS11-GFP + Arl8b + Myc-VPS18 + VPS41-HA (bottom row).
They were then fixed, permeabilized, and stained for the indicated markers. (top row) In control cells, hVPS11
exhibited a cytosolic staining pattern demonstrating little colocalization with lysosomal marker LAMP1.
(middle row) Overexpression of Arl8b resulted in a partial recruitment of hVPS11 to Arl8b+ lysosomes. Note
that a portion of VPS11 staining remained in Arl8b negative punctae. (bottom row) Overexpression of Arl8b,
VPS41, and VPS18 resulted in near total recruitment of VPS11 to lysosomes as evidenced by the triple
colocalization of VPS11, VPS41, and Arl8b staining. C. Arl8b and VPS41 promoted the recruitment of
VPS16 to lysosomes: HeLa cells were transfected with VPS16-GFP (top row), VPS16-GFP + Arl8b (middle
row), or VPS16-GFP + Arl8b + Myc-VPS18 + VPS41-HA (bottom row). They were then fixed, permeabilized,
and stained for the indicated markers. (top row) Similar to VPS11 (part B above), VPS16 localized
predominantly to the cytosol in control cells and did not display significant colocalization with lysosomal
markers. (middle row) Overexpression of Arl8b resulted in partial recruitment of VPS16 to lysosomes.
(bottom row) Overexpression of Arl8b, VPS41, and VPS18 resulted in near total recruitment of VPS16 to
lysosomes as evidenced by triple staining for VPS16, VPS41, and Arl8b. Higher magnification insets are shown
for clarity. All scale bars= 10 m.
Figure S6, related to Figure 7: LAMP1 acquisition on phagocytosed beads
depended on actin polymerization and occurred downstream of early
endosomes
A. LAMP1 acquisition depended on actin polymerization
B. Phagocytosed beads isolated from Arl8b silenced cells did not show a delay in
transferrin receptor acquisition
Figure S6: LAMP1 acquisition on phagocytosed beads depended on actin polymerization and
occurred downstream of early endosomes A. LAMP1 acquisition depended on actin
polymerization: To address the possibility that LAMP1 on phagocytosed latex beads was adsorbed
non-specifically during bead repurification we treated cells with the actin polymerization inhibitor
cytochalasin D. Since phagocytosis depends on actin reorganization drug treatment abolishes
phagocytosis. (left panel) Pre-treatment of RAW cells with 2 M cytochalasin D for 5 minutes
resulted in a dramatic reduction of LAMP1 acquisition on latex beads following 60 minutes
incubation. (right panel) The geometric MFI for LAMP1 on latex beads repurified 60 minutes after
addition for both control and Arl8b silenced RAW cells is plotted with or without cytochalasin D pre-
treatment. Note that in both Arl8b silenced and control cells drug treatment drastically reduced the
acquisition of LAMP1 on latex beads, indicating LAMP1 acquisition depended on actin
polymerization. This strongly suggests LAMP1 acquisition on beads was due to phagocytosis
followed by fusion with lysosomes. B. Phagocytosed beads isolated from Arl8b silenced cells did
not show a delay in transferrin receptor acquisition. In contrast to LAMP1, acquisition of
transferrin receptor on bead phagosomes isolated from Arl8b silenced cells was not delayed compared
to control cells (compare B to Figure 7B). There was a slight increase in transferrin receptor levels
detected on bead phagosomes isolated 30 minutes after infection in Arl8b silenced cells, suggesting a
connection between phagosomal maturation and acquisition of lysosomal markers (LAMP1) and loss
of early endocytic markers (transferrin receptor).
Supplemental Experimental Procedures
Table S1: Summary of shRNA sequences utilized in this study
21-mer
Name: mRNA target: Target Sequence: Source:
Arl8b-407*** hArl8b, NM_018184 AGGTAACGTCACAATAAAGAT TRC
Arl8b-346 hArl8b, NM_018184 CAGGTCAATTCAGTGAAGATA TRC
Arl8b-921 hArl8b, NM_018184 GCTGAAGATGAATATCCCTAA Cloned
Arl8b-377 hArl8b, NM_018184 AGTGGGCTTCAACATGAGGAA Cloned
Arl8b-602 hArl8b, NM_018184 GCTTGGAAACAAGAGAGATCT Cloned
Arl8b-347 hArl8b, NM_018184 AGGTCAATTCAGTGAAGATAT Cloned
Arl8b-404 mArl8b, NM_026011 CGAGGAGTCAATGCAATTGTT TRC
Arl8b-461** mArl8b, NM_026011 GCCTCTCGAAATGAACTGCAT TRC
Arl8a-1020 mArl8a, NM_026823 CCCTTTCAACACTCTGTTATT TRC
VPS11-2265 hVPS11, NM_021729 CCTCAAGCATATCGAGAACAA TRC
VPS16-1045 hVPS16, NM_022575 GCCAGCGAGGAAATCTTCAAA TRC
VPS18-708 hVPS18, NM_020857 CGTGAACCGAAATGGACAGAA TRC
VPS33A-1028 hVPS33A, NM_022916 CGAGGAAAGACACAATGCTAA TRC
VPS33A-277 hVPS33A, NM_022916 CGGCAGCTGATGTGAAGAATA TRC
VPS39-1883 hVPS39, NM_015289 CCTCGGCTTCTTAATAGAGAA TRC
VPS41-2134 hVPS41, NM_014396 CCATTGACAAACCACCATTTA TRC
MISSION* None CAACAAGATGAAGAGCACCAA Sigma
GFP-437* GFP TRC
Annexin A1* mAnnexinA1, NM_010730 GCTTTGGCAGATAAGTCTAAT TRC
Annexin A6* mAnnexinA1, NM_010730 CCTCTCTTCTTTGCTGATAAA TRC
Empty pLKO.1* None N/A TRC
All vectors listed as TRC are commercially available through Open Biosystems and Sigma. We are
happy to distribute vectors cloned in the lab.
*=Each of these “control” shRNA hairpins was used in almost every experiment and found to give
similar results. Additionally, results with transduction by these control shRNA were identical to non-
transduced (uninfected) cells in all experiments except where both controls are shown (Figure 1b/1c).
For clarity, all experiments in the manuscript utilize shMISSION transduced cells as the control unless
otherwise noted.
**=Figures targeting murine Arl8b utilize this hairpin unless otherwise noted
***=Figures targeting human Arl8b utilize this hairpin unless otherwise noted
Table S2: Summary of qPCR primers utilized in this study
Target mRNA molecule: Forward Primer Reverse Primer
hArl8b, NM_018184 CACCTTCGTCAATGTCATCG CCTATGTCCCAGATCTTTATTGTG
mArl8b, NM_026011 CTTCTGGAAGGAGGAGATGG TGTCCCAGATCTTTATTGTGAC
mArl8a, NM_026823 ATTGTGTATATGGTGGATGCT AGGGTGATGTCTATGTTGTC
hGapdh CATTTCCTGGTATGACAACGA GTCTACATGGCAACTGTGAG
m-actin GCTCTGGCTCCTAGCACCAT GCCACCGATCCACACCGCGT
Table S3: Summary of Antibodies utilized in this study
Presumed
Target molecule: Structures: Manufacturer, Cat. #/Clone:
LAMP1, human Lysosomes BD Biosciences 555798
LAMP1, murine Lysosomes BD Biosciences 553792
EEA1, human Early Endosomes BD Biosciences 610457
EEA1, murine Early Endosomes Calbiochem 324610
M-6-PR Golgi, Late Endosomes, Lysosomes Calbiochem 444105
Saposin C Golgi, Lysosomes Santa-Cruz, SC-27021
mCD1d Lysosomes, Cell surface 19G11, Gift from A. Bendelac
mCD1d Lysosomes, Cell surface 1B1, BD Biosciences 553843
CD1d•GalCer Lysosomes, Cell surface Gift from S. Porcelli
Actin (Phalloidin) Cytoskeleton Molecular Probes/Invitrogen
-Tubulin Microtubule filaments Sigma, B512
-Tubulin MTOC, Microtubule filaments Sigma, AK-15
COPI Median Golgi Gift from V. Hsu
Arl8b (antisera) Lysosomes Raised by YenZym Antibodies
CD63 Lysosomes BD Biosciences 556019
p150 Microtubule Filaments BD Biosciences 610473
Anti-HA Hemagglutinin tag Covance, MMS-101P
Expression constructs:
Arl8b expression constructs were obtained from Origene and cloned into pcDNA3.1 using the BamH1
and EcoR1 restriction sites. Point mutants were designed and obtained using Stratagene site directed
mutagenesis kits (Agilent). GFP-Rab7 and GFP-Rab9 expression constructs were obtained from Dr.
Steve Caplan. GFP-RILP and GFP-ORLP1 were obtained from Dr. Jacques Neefjes. VPS expression
constructs were obtained from Drs. J Wade Harper, Victor Faundez, and Robert Piper. All expression
constructs were sequence verified using standard vector primers. GST-Arl8b, GST-Arl8b mutants, and
GST-RILP were expressed from pGEX 4T-3 in BL-21 cells, purified using standard methods. His-
Arl8b was cloned in pet15b vector using NdeI and BamH1 restriction enzymes and retransformed in
BL-21 strain for protein production. Bacteria were induced with 0.5mM IPTG for 5 hours at 30
degrees C. After sonication, His-Arl8b was purified from the bacterial lysates using TALON cobalt
resin (Clontech). Purified protein was eluted from the beads by using elution buffer (Clontech,
HisTALON buffer set)
Lentiviral transduction:
For lentiviral transduction, U937 cells were plated in 96-well round bottom plates (10,000/well,
Corning) in polybrene (8 g/mL, SIGMA), and mixed with 10 L of viral supernatant (Day 0).
Puromycin was added after 24-48 hours at 2.5 g/mL for a minimum of three days to select
transductants. Monocyte-derived DC were transduced with lentivirus by spinoculation at 2000 rpm, 30
minutes at 25 degrees C (Day 2), puromycin added (2.0 g/mL, Day 3) and analyzed for CD1
presentation (Day 6-7). HeLa and RAW cells were plated at 100,000/well in 6-well plates (Corning) in
5 g/mL Polybrene and transduced by addition of 100 uL viral supernatant. 24-48 hours later,
puromycin was added at 2.5 g/mL (HeLa) and 5 g/mL (RAW) to select transductants and
experiments performed on Days 5-21 following transduction.
GST pulldowns:
GST, GST-Arl8b, and GST-Arl8b mutants were bound to GST beads. Lysates were prepared in 0.5%
CHAPS buffer containing GTPγS and were incubated with purified bound GST protein overnight
followed by washes with PBS or PBS containing GTPγS. Eluates were run on SDS-PAGE and
subjected to electrospray ionization tandem mass spectrometry (LC-MS/MS) to discover Arl8b
interacting proteins (Supplementary Figure 4) or were western blotted with anti-HA antibody
(Figure 4).
His-Arl8b was first incubated with 0.5mM GTPS in a nucleotide loading reaction. The GTP-loaded
Arl8b was then incubated with Talon resin for 30 mins at 4 degrees followed by blocking of beads with
5% BSA for 1 hour. Incubation of GST proteins : GST, GST-Vps41 (Novus biologicals) or GST-
RILP-Rab7 Binding domain was performed at 4 degrees C for 3 hours in 20mM Tris-HCl Ph 7.4 ,
150Mm NaCl, 1mM MgCl2, 5% Glycerol , 0.5% NP-40. After the incubations, beads were pelleted
by centrifugation (500g for 4 minutes at 4 degrees) followed by three washes in 25mM HEPES,
300mM NaCl, 90mM KCl, 1mM MgCl2, 0.2%Triton X-100 + 0.5mM GTPS and then analyzed by
SDS-PAGE. Western blotting was performed using anti-GST-HRP antibody (Millipore) to detect the
bound proteins. Coomassie brilliant blue and Silver stain were performed to detect His-Arl8b and
GST-tagged proteins respectively.
Yeast two hybrid analysis:
The Saccharomyces cerevisiae strain AH109 (Clontech) was maintained on Yeast extract, Peptone,
Dextrose (YPD) agar plates. Transformation was carried out by the lithium acetate procedure as
described in the instructions for the MATCHMAKER two-hybrid kit (Clontech). For colony growth
assays, AH109 cotransformants were streaked on plates lacking leucine and tryptophan and allowed to
grow at 30 degrees C for 5 days until colonies were large enough for further assays. An average of
three to four colonies was then chosen and suspended in water, equilibrated to the same optical density
of 600 nm and replated on plates lacking leucine and tryptophan (+HIS) as well as plates also lacking
histidine (−HIS).
Arl8b antisera and western blotting
To generate Arl8b antisera, peptide was conjugated to KLH as a protein carrier, injected into rabbit,
and boosted bi-weekly for 6 weeks. Rabbit IgG was then affinity purified against peptide. Lysates
from 107 U937 or 107 RAW cell equivalents/lane were made in 0.5% Triton X-100 and analyzed on
15% SDS-PAGE, transferred to PVDF membranes (Bio-Rad) and blotted with Rabbit anti-Arl8b IgG
(1 ug/mL) followed by donkey-anti rabbit HRP (Jackson Labs) at 1:30,000 dilution. Membranes were
stripped and reprobed with anti-human -actin (Abcam) for human cell lysates or anti-murine GAPDH
(Abcam) followed by appropriate secondary HRP conjugates (Jackson Labs). All blots were
developed with ECL reagents and exposed to film
Colocalization analysis using MetaMorph Imaging software:
Images from control or Arl8b silenced cells were acquired using identical laser power and gain settings
on a single day within an experiment. Each channel (R,G, B) was considered separately with each
pixel assigned an intensity value 0-255 with great care taken to assure that no pixel was saturated in
any image used for quantification (no values = 255). A lower threshold is set whereby areas of the
image not containing cells or staining are excluded, and the same threshold was always applied to
control and Arl8b silenced cells. This value was 10-30 depending on the stain and never excluded
visibly positive regions. The software then calculates the area overlapping between two channels
where both stains are positive over the lower threshold and reports it in percentage terms. This
analysis was utilized to give the reported co-localization values in Figures 2, 5, 7, and Supplementary
Figure 2.
Phagosomal Assays
3 m latex beads were purchased (Polysciences Inc) and coated by rotation with .4 mg/mL murine IgG
(Sigma) for 1 hour at 37°C and added to RAW cells at a ratio of 10 beads/cell. Bead/cell mixtures
were centrifuged at 300g to synchronize uptake. Following incubation for indicated times at 37°C,
coverslips were washed 3x in complete DMEM to remove unbound beads, and fixed and stained as
described. E. coli (EPEC E1348/69, gift from Lynn Bry, Harvard University) were added to RAW
cells at multiplicity of infection (MOI) of 20 and centrifuged at 300g for 5 mins. The mix was then
incubated for 60 mins at 37°C, 5% CO2 followed by treatment with 100 g/mL gentamycin and chase
for the indicated times. For microscopy, cells were fixed and stained for LAMP1 and murine IgG. For
CFU, cells were lysed in 0.2 % Triton in PBS for 5 mins and plated in serial dilution on LB agar plates.
Following 24 hours, CFU was obtained by analysis of colonies and surviving E. coli per RAW cell
backcalculated.