Anatomy and Regenerative Biology Faculty Publications Anatomy and Regenerative Biology Simultaneous Measurement of Smoothened Entry into and Exit from the Primary Cilium Recommended Citation Simultaneous Measurement of Smoothened Entry into and Exit from the Primary Cilium

Ciliary accumulation of signaling proteins must result from a rate of ciliary entry that exceeds ciliary exit, but approaches for distinguishing ciliary entry vs. exit are lacking. Using a photoconvertible fluorescent protein tag, we establish an assay that allows a separate but simultaneous examination of ciliary entry and exit of the Hedgehog signaling protein Smoothened in individual cells. We show that KAAD-cyclopamine selectively blocks entry, whereas ciliobrevin interferes initially with exit and eventually with both entry and exit of ciliary Smoothened. Our study provides an approach to understanding regulation of ciliary entry vs. exit of Hedgehog signaling components as well as other ciliary proteins. Copyright: ß 2014 Kim et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Data Availability: The authors confirm that all data underlying the findings are fully available without restriction. All relevant data are within the paper and its Supporting Information files.A.B. is an investigator of the Howard Hughes Medical Institute. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.


logy and Rege
erative Medicine
Stanford University School of Medicine
StanfordCAUnited States of America

Department of Anatomy and Regenerative Biology
George Washington University School of Medicine and Health Sciences
WashingtonDCUnited States of America


Indiana University School of Medicine
United States of America

Simultaneous Measurement of Smoothened Entry Into and Exit From the Primary Cilium
August 13, 20144F6E280780B6BA9C46726979CF5412C710.1371/journal.pone.0104070Received March 21, 2014; Accepted July 5, 2014;
Ciliary accumulation of signaling proteins must result from a rate of ciliary entry that exceeds ciliary exit, but approaches for distinguishing ciliary entry vs. exit are lacking.Using a photoconvertible fluorescent protein tag, we establish an assay that allows a separate but simultaneous examination of ciliary entry and exit of the Hedgehog signaling protein Smoothened in individual cells.We show that KAAD-cyclopamine selectively blocks entry, whereas ciliobrevin interferes initially with exit and eventually with both entry and exit of ciliary Smoothened.Our study provides an approach to understanding regulation of ciliary entry vs exit of Hedgehog signaling components as well as other ciliary proteins.

Introduction

The primary cilium, a microtubule-based organelle about 5 mm in length and 200 nm in diameter, projects from the surface of vertebrate cells to sense and interpret a variety of extracellular signals [1].Previous work has suggested that the dynamic movement of receptors and other proteins into and out of cilia regulates the activity of signaling complexes that ultimately trigger responses in the cell [2].It has been proposed that ciliary accumulation of signaling proteins results from a rate of ciliary entry that exceeds the rate of ciliary exit [3].An understanding of signaling receptor trafficking into cilia is now emerging [4]; however, the mechanisms that underlie membrane protein removal from cilia and the regulation of this trafficking step remain largely unexplored.The major challenges arise not just from the small size of the primary cilium, but also from the difficulty of separately evaluating ciliary entry and exit of proteins that traffic through the primary cilium.In this context, development of new assays that can distinguish ciliary entry and exit are crucial to progress in understanding regulation of ciliary trafficking.

The Hedgehog (Hh) signaling pathway organizes pattern formation in a variety of embryonic tissues and functions postembryonically in homeostatic processes.Hh pathway dysfunction thus can lead to birth defects such as holoprosencephaly (HPE) [5] or proliferative disorders such as the growth of malignant tumors [6].The quiescent state of the Hh signaling pathway is maintained by Patched (Ptc) inhibition of Smoothened (Smo) [7].This inhibition is lifted by binding of the extracellular Hh protein signal to Ptc, thus unleashing Smo activity and initiating a series of intracellular events that lead to changes in gene transcription.Recent studies have highlighted the importance of the primary cilium in transduction of mammalian Hh signals.Smo and other Hh pathway components in mammalian cells traffic through the primary cilium and accumulate upon Hh stimulation and Ptc inactivation [8][9][10][11][12][13].Small molecules that either activate or inactivate Smo can also modulate signaling activity and ciliary localization of Smo [3,14].Given that accumulation of Smo in the primary cilium is one of the earliest hallmarks of Hh pathway activation, understanding regulation of Hh signal transduction depends critically on unveiling the molecular mechanism of Smo accumulation in the primary cilium.

Whereas ciliary entry has been emphasized as a critical point of regulation, the findings that Smo continuously shuttles into and ou

of the cilium in unstim
lated cells [11,15] and that levels of ciliary Smo eventually decrease once stimulation is terminated leave open the possibility that either entry or exit rates could be the target for regulation upon pathway engagement.Establishment of assays to separately monitor Smo ciliary entry and exit therefore could illuminate the general mechanism underlying protein accumulation in the primary cilium as well as shed light into how the Hh signaling pathway is regulated.

Here, we fuse a ph toconvertible fluorescent protein, mEos2 [16], to the C-terminus of Smo (Smo-mEos2) and establish a livecell imaging assay that allows a simultaneous examination of ciliary entry and exit of Smo in individual cells.Using this assay, we find that activated Smo enters and exits the cilium continuously with a ciliary retention half-life of approximately two hours.We also find that the small molecule KAAD-cyclopamine selectively blocks ciliary entry of Smo, whereas long-term exposure to ciliobrevin eventually interferes with both the ciliary entry and exit of Smo.Our study provides an approach to understanding the separate regulation of ciliary entry vs. exit of signaling proteins within and beyond the Hh pathway.


Results and Discussion


Ciliary entry and exit of Smo-mEos2 in SAG-treated cells

To simultaneously mon tor how the Hh pathway protein Smo enters and exits the primary cilium, we have fused the photoconvertible fluorescent protein mEos2 [16] to the Cterminus of Smo, which permits normal function of the resulting chimeric Smo-mEos2 protein upon introduction into Smo 2/2 cells (Fig. S1).We generated an NIH 3T3 cell line stably expressing the Smo-mEos2 protein (NIH 3T3/Smo-mEos2), in which the Smo-mEos2 accumulated within cilia in response to Shh or SAG, a small molecule that binds directly to Smo (Fig. 1a-d).

The NIH 3T3/Smo-mEos2 cell line shows comparable response to ShhN or SAG as in NIH 3T3 cells (Fig. 1e).

We first treated the NIH 3T3/Smo-mEos2 cells with SAG for 24 hours to allow Smo-mEos2 to accumulate in primary cilia.The cilium highlighted by Smo-mEos2 was selected as a region-ofinterest (ROI) for photoconversion with a 405-nm laser.Upon photoconversion, mEos2 shifts from a peak excitation at 506 nm and peak emission at 519 nm (Smo-mEos2 native, hereafter Smo-mEos2 N ), similar to GFP, to a peak excitation at 573 nm and peak emission at 584 nm (Sm

mEos2 photoconverted, hereafter Smo-mEos2 P ), similar to RFP.B
th Smo-mEos2 N and Smo-mEos2 P were subsequently monitored through time-lapse imaging.Thus, an increase of green fluorescence within the cilium reflects Smo-mEos2 N entering from the cell body, while a decrease of red fluorescence in the cilium reveals Smo-mEos2 P exiting the cilium.

In previous photobleaching experiments, a significant recovery of ciliary fluorescence for the intracellular transport protein IFT88 was detected several minutes after photobleaching [17].In contrast, we detected little or no recovery of Smo-mEos2 N (ciliary entry) 10 minutes after photoconversion (Fig. S2).Similarly, no noticeable reduction of Smo-mEos2 P (ciliary exit) was detected within a similar time frame (Fig. S2).

This delay in recovery of ciliary Smo-mEos2 N relative to IFT88 may arise from the slower kinetics of Smo entry into the cilium, whether by limited lateral diffusion [18] of membrane proteins from the plasma membrane into the primary cilium, proposed to result from a diffusion barrier at its base [17], or alternatively from the relatively slow process of entry via fusion of internal vesicles [19].We therefore increased the time-lapse imaging interval and first detected an increased level of Smo-mEos2 N in the cilium around 25 minutes after photoconversion.A slight decrease of Smo-mEos2 P was also noticed within a similar interval.Using this assay, we observed that Smo-mEos2 proteins constantly enter and exit the cilium in the presence of SAG, with a ciliary retention half-life of approximately 2 hours (Fig. 2).

Given that neither electron microscopic images nor proteome analyses have revealed

e presence of proteasomes in cilia [20][21][22], it is unlikely
hat the observed decrease of Smo-mEos2 P is due to its degradation within the primary cilium.We further examined the possible role of protein degradation in turnover of Smo-mEos2 by measuring the stability of Smo-mEos2 protein in cells treated with cycloheximide to block new protein synthesis.We found that Smo-mEos2 in the presence of SAG has a half-life of more than 6 hours (Fig. S3), three-fold longer than the approximate 2-hour ciliary retention half-life of Smo-mEos2 in SAG-treated cells.Therefore, exit of Smo-mEos2 accounts for most, if not all, of its turnover in the primary cilium of SAG-treated cells.Interestingly, we measured a half-life of total Smo-mEos2 protein on the order of 2 hours in cells not treated with SAG (Fig. S3), suggesting that SAG binding to Smo may increase its stability, which is consistent with previous studies showing that SAG-binding to Smo may aid in its maturation [23].

Ciliobrevin A reduced both ciliary entry and exit of Smo-mEos2

It was previously reported that inactive Smo accumulated in primary cilia of cells with disrupted retrograde intraflagellar transport [11,15].We reproduced this ciliary accumulation of Smo in the absence of Hh ligands or Smo agonists in NIH 3T3/Smo-mEos2 cells by adding a low level of ciliobrevin A (CBA) [24], a specific small molecule inhibitor of the minus end-directed microtubule motor cytoplasmic dynein.After 24 hours of treatment with 10 mM CBA, the proportion of ciliated cells was dramatically reduced to about 30% of the entire cell population (Fig. S4C).Smo-mEos2 was detected in more than 70% of these ciliated cells (Fig. S4A, B, D).

To assay the ciliary entry and exit kinetics of inactive Smo caused by disrupted retrograde intraflagellar transport, we treated the NIH 3T3/Smo-mEos2 cells with 10 mM CBA for 24 hours to allow Smo-mEos2 to accumulate in primary cilia.The cilium highlighted by Smo-mEos2 N was selected for photoconversion and both the native and the photoconverted species of Smo-mEos2 were subsequently monitored through time-lapse imaging.When compared with SAG-treated cells, we noticed a much slower entry and exit of both native and photoconverted Smo-mEos2 in the cilium of cells treated with CBA (Fig. 3).Specifically, at 200 minutes after photoconversion, about 80% of Smo-mEos2 P remained in the cilium of CBA-treated cells, whereas less than 20% was left in the cilium of SAG-treated cells.Similarly, compared to more than 80% recovery of Smo-mEos2 N in the cilium of cells treated with SAG, about 20% of Smo-mEos2 N was detected in the cilium of CBA-treated cells.

These results suggest that defects in retrograde intraflagellar transport dramatically slowed down both ciliary entry and exit of Smo.A previous study has demonstrated that ciliobrevins are specific small molecule inhibitors of cytoplasmic dynein, and have no significant effect on kinesin-dependent anterograde microtubule sliding even at a concentration 10 times higher than was used in our experiment [24].The reduced ciliary exit rate of Smo could be explained as a direct result of the defects in retrograde intraflagellar transport caused by CBA.The reduced ciliary entry rate of Smo in contrast is likely to be an indirect effect of CBA resulting from a cumulative general disruption of ciliary trafficking over time; initial entry is unlikely to be affected, as increase in ciliary Smo was noted when the NIH 3T3/Smo-mEos2 cells were pretreated with SAG for 24 hours, followed by photoconversion and time-lapse imaging immediately after introducing CBA (Fig. S5).


KAAD-cyclopamine selectively reduces ciliary entry of Smo-mEos2

Ciliary accumulation of Smo has been reported to occur upon treatment with cyclopamine [3]

uriously
although we and others have confirmed this effect [11,19,25], we failed to see Smo accumulation upon treatment with KAAD-cyclopamine (data not shown), a highly potent derivative of cyclopamine [23,26].The different effects of cyclopamine and KAAD-cyclopamine on Smo ciliary localization were also found when using our NIH 3T3/ Smo-mEos2 cells (Fig. S6).Furthermore, we found in a set of timeseries experiments that treatment with 300 nM KAAD-cyclopamine reversed the ciliary accumulation of Smo-mEos2 that

induced by
pre-treatment with 100 nM SAG.Specifically, we started to detect reduction of ciliary Smo-mEos2 about 4 hours after adding KAA

cyclopa
ine, and ciliary Smo decreased by more than 90% within an additional 8 hours (Fig. 4).To rule out the possibility that the KAAD-cyclopamine-associated removal of ciliary Smo-mEos2 is an artifact due to the mEos2 tag on Smo, we tested the effect of KAAD-cyclopamine on endogenous Smo in NIH 3T3 cells, and found that KAAD-cyclopamine eliminated SAG-induced ciliary accumulation of endogenous Smo (Fig. S7).These observations indicate that the effect of KAAD-cyclopamine on Smo-mEos2 indeed reflects the behavior of wild type Smo proteins.

The observed reversal of ciliary accumulation of Smo-Eos2 could be explained in several ways: (i) KAAD-cyclopamine-induced degrada

on of Smo; (ii) decreased ciliary entry rate o
Smo; or (iii) accelerated ciliary exit of Smo.To distinguish these possibilities, we first used cycloheximide chase experiments to compare the stability of total Smo-mEos2 protein levels in cells treated with SAG vs. SAG followed by addition of KAADcyclopamine.Adding KAAD-cyclopamine in cells pre-treated with SAG did not decrease the half-life of Smo-mEos2 (Fig. S8), and instead caused a slight

crease in total Smo-
Eos2 protein levels as compared to cells treated with SAG alone.We therefore ruled out a destabilizing effect of KAAD-cyclopamine as the cause of Smo-mEos2 ciliary loss.

To test for KAAD-cyclopamine triggered changes on ciliary trafficking of Smo-mEos2 we used the photoconversion assay.NIH 3T3/Smo-mEos2 cells were pre-treated with 100 nM SAG for 24 hou

to allow Smo-mEos2 to accumulate in primary cilia, and 300
M KAAD-cyclopamine was added to cells followed by photoconversion of ciliary Smo-mEos2.Both Smo-mEos2 N and Smo-mEos2 P were subsequently monitored through time-lapse imaging.About 200 minutes after photoconversion, we observed a dramatic reduction of Smo-mEos2 P in the cilium of cells exposed to KAAD-cyclopamine, which is similar to the ciliary exit rate of Smo-mEos2 in the cilium of cells treated with SAG alone.In contrast to the normal ciliary exit rate indicated by Smo-mEos2 P , the ciliary entry rate indicated by recovery of Smo-mEos2 N in the cilium of KAAD-cyclopamine treated cells was much slower: thus, at the time when Smo-mEos2 N was almost completely recovered in the cilium of SAG treated cells, less than 20% recovery was achieved by Smo-mEos2 N in cells treated with SAG and KAADcyclopamine (Fig. 5).These data thus suggest that KAADcyclopamine-induced loss of Smo from the cilium of SAG treated cells is due to decreased ciliary entry rather than accelerated ciliary exit.Additionally, KAAD-cyclopamine blocked the ciliary accumulation of Smo-mEos2 triggered by CBA (Fig. S9).This result supports the conclusion that KAAD-cyclopamine decreases ciliary entry of Smo-mEos2 even in the absence of agonists.In addition, the finding that blocked ciliary entry of Smo by KAADcyclopamine prevents the accumulation triggered by treatment with CBA alone reinforces our conclusion that CBA does not initially block ciliary entry of Smo, and that the entry block we measured after 24 hours of treatment with CBA indeed results from a cumulative overall defect in ciliary trafficking.

Ott and Lippincott-Schwartz [27] recently showed that photoconversion could be used to highlight and track the movement of a subset of molecules within the primary cilium, and Ye et al. tracked the movement of single molecules within the cilium [28].In both of these studies the major focus was on tracking individual molecules or complexes within the cilium by photoconverting or labeling a small subset of the molecules.Our study differs in that we completely photoconverted a ciliary protein, Smo, in order to measure the ensemble average of ciliary entry and exit kinetics.This required rel