In This IssueAuthor(s): Nicole LeBrasseurSource: The Journal of Cell Biology, Vol. 165, No. 2 (Apr. 26, 2004), pp. 162-163Published by: The Rockefeller University PressStable URL: http://www.jstor.org/stable/1622180 .

Accessed: 25/06/2014 07:38

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .http://www.jstor.org/page/info/about/policies/terms.jsp

.JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range ofcontent in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new formsof scholarship. For more information about JSTOR, please contact support@jstor.org.

.

The Rockefeller University Press is collaborating with JSTOR to digitize, preserve and extend access to TheJournal of Cell Biology.

http://www.jstor.org

This content downloaded from 185.2.32.14 on Wed, 25 Jun 2014 07:38:06 AMAll use subject to JSTOR Terms and Conditions

In This Issue

F~ ~? ???

?;,.? :?.

.i?i. '-I :: i?:

Mitochondria (green) are fragmented if I-Mgml and s-Mgml levels are uneven (right).

Lethargic mitochondria are left out T he proper processing ofa mitochondrial fusion protein called Mgml depends

on sufficient ATP, based on results from Herlan et al. (page 167). Mgm l's energy dependence may ensure that tired mitochondria get left behind.

Mgml can be found in the mitochondrial intermembrane space in two forms, the shorter of which (s-Mgm 1) is generated by protease cleavage of the long form

(l-Mgm 1). The authors find that a balance between the two forms maintains mitochondrial morphology-deviation from a 1:1 ratio in either direction causes fragmentation. Although little is known about Mgml function, this need for balance might be explained if s- and l-Mgml work as heterodimers.

Balanced production of the two forms depends on a hydrophobic stretch in l-Mgm 1 following the targeting sequence. Mutations that further increased its hydrophobicity favored production of the long form, whereas a decrease in hydrophobicity favored the short form. This suggests that hydrophobic-dependent lateral movement of I-Mgm 1 out of the import channel into the inner membrane prevents its cleavage.

But if the protein instead translocates further into the mitochondrial import channel, s-Mgml is formed. This translocation depends on the protein import motor (including the ATP-dependent chaperone Sscl), which drives the NH2 terminus of Mgm 1 into the matrix until a second hydrophobic domain reaches the inner membrane. There, the second domain is cleaved by the Pcpl protease to produce s-Mgml.

Cells impaired in ATP synthesis made mostly I-Mgml and had a fragmented mitochondrial network. The need for energy during Mgml import might ensure that mitochondria that are inept in ATP production, perhaps due to oxidative damage, are excluded from the network and are thus not inherited by daughter cells. I

Arpless comets are fascin-ating A

ctin elongation is a force to be reckoned with. Results on page

233 by Brieher et al. show that elongating bundles of actin can, by themselves, propel a bacterium, and do so faster than frequent actin nucleation does.

Actin filament nucleation by Arp2/3 is a prerequisite for movement of the intracellular pathogen Listeria monocytogenes. Through repeated nucleations, Arp2/3 builds a comet tail of short branched actin filaments that push the bug from behind. But Brieher and colleagues show that once some nucleation occurs, comet tails can be built without Arp2/3.

Upon Arp2/3 inhibition, the branched filaments elongated into long barrel- shaped tails encircling the bacterium. Bugs building these structures were faster than those using Arp2/3. As the barrels are not attached at the bug's rear, it is not clear how movement is achieved. Fractionation experiments

showed that the acti n-bundling protein fascin is required, although other bundling proteins also worked. The barrels might work like railroad tracks, with the bundling proteins preventing buckling of the filaments. No motor for these tracks has been found at the actin-bacterium interface however.

If there is no motor, motion may rely on diffusion followed by capture by unbundled filaments. If the more posterior, bundled actin is unable to bind to the bug's surface, net for- ward movement might occur. At least one other cytoplasmic factor was required in addition to fascin for Arpless motility. Its identification should supply clues regarding the force-generating mechanism.

In infected cells, the bug probably uses both polymerization methods, as both barrels and branched filaments were seen by EM. The combination might be needed to raise enough force to push past the host plasma membrane

... ......... . .iil~l~~:::j .:illliijiiijilj iL ~l ?. ? ::::!:: ?:i. ?: ?. ::illiiiii;'i)N H ,fijiiillili: :::::::::i? ;?::;::::::::::: OR,? :::I 5;:::. ? :.fi.V:: :: . . .... .......?: -...... .....

y::t.

.. ....... .:::: ?:: ........ . .....: ?r::::I jliii: ........ ....... : ??? r~iii:'l: :' " "' ::' :.i:

'tt ftf. t~tiIIP1;Z1.?' l :: .

i~t:1~:.. ............ ........

as the pathogen infects other cells. In other systems, such as fibroblasts, a switch from Arp2/3- to elongation- based assembly might initiate the transition from lamellipodia, which are rich in branched filaments, to filo- podia, which contain fascin and parallel actin bundles. I

162 The Journal of Cell Biology I Volume 165, Number 2, 2004

This content downloaded from 185.2.32.14 on Wed, 25 Jun 2014 07:38:06 AMAll use subject to JSTOR Terms and Conditions

TEXT BY NICOLE LEBRASSEUR LEBRASN@ROCKEFELLER.EDU

PKCs pick through Ca2+

Calcium signals destined for mitochondria

are first screened by protein kinase C (PKC) isoforms, according to Pinton et al., on page 223. Different PKCs then adjust the organelles' responses to their liking.

PKC is a Ca2+-activated kinase that comes in many flavors. Pinton et al. analyzed the contribution of several of these flavors to Ca2 2

signaling by overexpressing or inhibiting individual isoforms in cells. Their effects were assessed by stimulating cells with an extra- cellular agonist to elicit Ca2+ stores from the ER into the cytoplasm.

One flavor, PKCa, limited all Ca2+ signaling by dampening this ER release. Others, however, worked specifically at the mitochondria. PKCC increased mitochondrial Ca2+ uptake,

Histamine Histamine 100--

80-

40. 20.

0 -

PKC3 Inhibitor

Histamine Histamine

100

80.-.

• 60- 40- 20-

0 .

Mitochondrial desensitization to the second of two stimuli is partly lost if PKCP is inhibited (bottom).

whereas PKC3 and 8 reduced it. Thus, activation of PKC upon initial stimulation may alter the effect of subsequent Ca2+ spikes. Mitochondria are known to respond less to the second of two consecutive stimuli. The authors now show that PKC3 is needed for this dampening. This function might make cells less sensitive to apoptotic signals, for example, which induce mitochondrial Ca2" influx.

The PKC isoforms that decrease Ca2" uptake also decrease mitochondrial mem- brane potential. There may be a causal link, but mitochondrial Ca2" transporters have not been cloned, so identifying the relevant PKC substrates will require the purification of mitochondrial proteins that are phosphor- ylated in response to Ca2+.

Next vesicle SNAP'd up S ecretory vesicles wishing to play follow-the-leader use

a particular SNARE to show them where the leader

went, as shown by Takahashi et al. on page 255. Vesicles heading to the plasma membrane sometimes

selectively bind to another that has already fused there. This process, called sequential exocytosis, is efficient at large-scale secretion because vesicles in the cytosol can be mobilized without being transported all the way to the

plasma membrane. Using 2-photon imaging, the authors show that sequential exocytosis is directed by a plasma membrane SNARE called SNAP25.

Sequential exocytosis was examined in insulin-secreting pancreatic 3 cells, in which SNAP25 diffused into the

bulge where the leading exiting vesicle had fused with the plasma membrane. SNAP25 rarely diffused into

spots where only one vesicle exited, but was seen at the

majority of sequential exocytosis sites.

Although sequential exocytosis is common in many exocrine or endocrine cell

types (which contain

SNAP25), it accounted for only a small fraction of p cell exocytosis. Cholesterol depletion

SNAP25 (red) moves into the bulge (outlined) where sequential exocytosis is taking place (left to right).

freed SNAP25 for easier diffusion and increased

sequential exocytosis several fold. Restricting SNAP25 to lipid rafts might thus be one way to prevent exhaust-

ing insulin reserves, which are doled out steadily in small quantities. M

Tracking transcripts

Mobile mRNAs rebrighten (left to right) a bleached speckle (circled).

m RNAs do not travel willy-nilly through the nucleus, according to Molenaar et al. (page 191). Instead, they

are transported by an energy-dependent mechanism that may bring them to quality control sites before they are exported to the cytoplasm.

Random diffusion from transcription sites to nuclear pores was generally accepted as the travel mode of preference for polyadenylated mRNAs. Based on the high mobility of oligo(dT) probes, transcripts were assumed to be moving through the nucleoplasm at rates comparable to diffusion. But Molenaar and colleagues find that these speeds were probably over- estimates resulting from free probe. Using a tighter-binding oligo(U) probe, they find that mRNA moves 10-fold more slowly than previous estimates.

This movement is energy dependent, indicating that an active process transports the mRNAs, perhaps by a motor or along chromatin fibers. The group now plans to inhibit nuclear mRNA- binding proteins to identify those that are essential for transport.

Transcripts were mobile even at speckles-putative nuclear storage sites for RNA-processing enzymes. This dynamic associa- tion suggests that transcripts are not important structural elements of speckles. Most transcripts passed through speckles at least briefly, so transcripts might instead be sent there to be checked for proper splicing. Since speckles contain splicing factors, they might even fix mRNAs that fail inspection, thus accounting for the small fraction of transcripts that were immobile in speckles. N

In This Issue 163

This content downloaded from 185.2.32.14 on Wed, 25 Jun 2014 07:38:06 AMAll use subject to JSTOR Terms and Conditions