Abstract The ability of cells to receive, process, andrespond to information is essential for a variety of biologi-cal processes. This is true for the simplest single cell entityas it is for the highly specialized cells of multicellularorganisms. In the latter, most cells do not exist as indepen-dent units, but are organized into specialized tissues.Within these functional assemblies, cells communicate witheach other in diVerent ways to coordinate physiologicalprocesses. Recently, a new type of cell-to-cell communica-tion was discovered, based on de novo formation of mem-branous nanotubes between cells. These F-actin-richstructures, referred to as tunneling nanotubes (TNT), wereshown to mediate membrane continuity between connectedcells and facilitate the intercellular transport of variouscellular components. The subsequent identiWcation of TNT-like structures in numerous cell types revealed some struc-tural diversity. At the same time it emerged that the directtransfer of cargo between cells is a common functionalproperty, suggesting a general role of TNT-like structuresin selective, long-range cell-to-cell communication. Due tothe growing number of documented thin and long cell pro-trusions in tissue implicated in cell-to-cell signaling, it isintriguing to speculate that TNT-like structures also existin vivo and participate in important physiological pro-cesses.
AbbreviationsTNT Tunneling nanotubePD PlasmodesmataPC12 Rat pheochromocytomaNRK Normal rat kidneyDC Dendritic cellsEPC Endothelial progenitor cellsCM Cardiomyocytes(E)GFP (Enhanced) Green Xuorescent proteinEGFP-f Farnesylation signal of c-Ha-Ras fused to the
C-terminus of EGFPGPI-GFP Glycosylphosphatidylinositol conjugated to GFPMHC Major histocompatibility complexHIV-1 Human immunodeWciency virus type 1MLV Murine leukemia virus
Introduction
Communication with the environment is a basic principleof any biological system. With the increasing complexity ofhigher organisms, cells had to evolve diverse mechanismsto exchange spatial and temporal information crucial fortissue organization and the maintenance of the organism asa whole. Certainly, tissues are not just a loose accumulationof individual cells, but a highly organized population ofinteracting cells. Thus, in the various types of animal andplant tissues, cell-to-cell communication is important formaintaining a supracellular organization. Probably the best-characterized example of such an organization is given bythe animal central nervous system, which is composed of acomplex network of interconnected neuron and glial cells.It is widely accepted that the complexity of the brain reX-ects its enormous number of intercellular links provided bysynaptic connectivity (including gap junctions) between
S. Gurke · J. F. V. Barroso · H.-H. Gerdes (&)Department of Biomedicine, University of Bergen, Jonas Lies vei 91, 5009 Bergen, Norwaye-mail: [email protected]
123
540 Histochem Cell Biol (2008) 129:539–550
axons and dendrites. Gap junctions, through which mole-cules <1 kDa can penetrate, are also important players inestablishing supracellular organization in other tissues con-taining electrically excitable cells as heart and smooth mus-cle cells. Furthermore, these proteinaceous channelsinterconnect cells of the multicellular layer forming epithe-lial tissues. Finally, cell-to-cell coupling via gap junctionsis a general phenomenon during early embryogenesis,where most cells are electrically coupled. Perhaps due tothe presence of a thick cell wall, plant cells have evolvedthin membrane channels referred to as plasmodesmata (PD)to allow direct connections over longer distances (Balunkaet al. 2004a; Cilia and Jackson 2004; Gallagher and Benfey2005). Because these bridges provide both membrane andcytoplasmic connectivity between cells, plants are regardedas a supracellular assembly (Balunka et al. 2004b, c). PD, inaddition to the functions known for gap junctions in ani-mals, facilitate the transfer of ribonucleoparticles, tran-scription factors and viruses (Ruiz-Medrano et al. 2004).
In 2004, a new principle of cell-to-cell communicationbetween animal cells, based on the formation of thin mem-brane channels, was reported (Rustom et al. 2004). Thesechannels, referred to as tunneling nanotubes (TNT), wereinitially found in cultures of rat pheochromocytoma (PC12)cells. As PD, they were shown to mediate membrane conti-nuity between connected cells. TNT permit the direct inter-cellular transfer of organelles, cytoplasmic molecules, andmembrane components (Gerdes et al. 2007). Subsequentstudies on other cell types revealed morphologically similarstructures directing intercellular transfer of cargo, includingpathogens. It thus becomes apparent that nanotubularbridges provide an important and general mechanism ofcell-to-cell communication between animal cells. In the fol-lowing, we summarize the current knowledge and develop-ment in the Weld of nanotubular communication. As
awareness of the potential physiological implications ofthese structures, we will give a short overview on publisheddata describing long and thin cellular protrusions in tissueimplicated in cellular communication.
Diversity of TNT-like structures
TNT connecting PC12 cells are typically 50–200 nm indiameter and can reach lengths up to several cell diameters.They are stretched, interconnecting cells at their nearestdistance (Fig. 1) (Hodneland et al. 2006; Rustom et al.2004). Their structural integrity is sensitive to mechanicalstress, chemical Wxation, and even to prolonged light expo-sure. The feature that most strikingly distinguishes TNTfrom other cellular protrusions is that they hover in themedium and have no contact to the substratum (Fig. 1a, a1).A structural characterization of TNT between PC12 cellsrevealed that they contain F-actin as a prominent cytoskele-ton element but no microtubules (Table 1). Further charac-terization at the ultra-structural level showed a seamlesstransition of the surface membrane of the TNT with theplasma membranes of the connected cells (Fig. 1b, b1, b2).In addition, a continuous membrane with both connectedcells was evident in transmission electron micrographs(Fig. 2a, a1, a2) (Table 1). These observations, together withthe documentation of a limited lateral diVusion of Xuores-cent membrane proteins occurring between the plasmamembranes of TNT-connected cells, led to the proposal thatTNT mediate intercellular membrane continuity (Fig. 3a)(Gerdes et al. 2007; Rustom et al. 2004).
Based on the morphological criteria deWned for PC12cells, similar TNT-like connections were subsequently iden-tiWed for several permanent cell lines and primary cultures(for a comprehensive overview see (Gerdes et al. 2007))
Fig. 1 Architecture of TNT between cultured PC12 cells. (a) 3D Xuo-rescence image ((x¡y)-maximum projection of 40 consecutive 400 nmsections) of a wheat germ agglutinin-stained TNT connecting two livePC12 cells. (a1) (x¡z)-projection in the plane of the TNT indicated in(a). (b) Scanning electron micrograph (SEM) showing the ultra-struc-
ture of a TNT between two PC12 cells. The boxed areas are shown ashigher magniWcation images (b1, b2). ModiWed from Rustom et al.(2004) Science 303:1007–1010. Scale bars, a, a1, b, 5 �m; b1, b2,500 nm
123
Histochem Cell Biol (2008) 129:539–550 541
(Table 1). TNT-like bridges between dendritic cells (DC)(Watkins and Salter 2005) and perhaps also those connectingneonatal rat cardiomyocytes (CM) and adult human endothe-lial progenitor cells (EPC) (Koyanagi et al. 2005), appear tobe most closely related to the TNT described for PC12 cells.The Xow of cytoplasmic molecules suggests membrane con-tinuity between connected cells (Table 1). This contrasts theTNT-like bridges between T cells, which have been charac-terized lately in considerable detail (Sowinski et al. 2008).The observed T-cell nanotubes have an average length of»20 �m, a diameter of 180–380 nm, and are not tethered tothe substratum. Moreover, as in PC12 cells, only F-actin but
not microtubules were detected in T-cell nanotubes (Fig. 3).However, despite the overall similarity in architecture andcapacity to facilitate intercellular transfer of cargo betweennanotubes of PC12 cells, DC, CM/EPC and T-cells, the latterdo not mediate membrane continuity between connectedcells (Fig. 2b, b1, b2) (Table 1). This was concluded from theobservation that nanotubes between T-cells (i) do not facili-tate the transfer of cytoplasmic molecules, (ii) do not permitfree diVusion of Xuorescent plasma membrane componentsand (iii) display a junctional border between the nanotubeand the connected T-cell at the ultra-structural level(Fig. 2b1). Accordingly, they were classiWed as “not open-
Table 1 TNT-like structures in vitroa
a Only those publications that fulWll at least one of the listed criteria are shownb Accessed by electron microscopyc Gurke, S., Barroso, J., Bukoreshtliev, N., Gerdes, H.-H., unpublished datad These molecules were shown to localize in TNT-like structures, but their intercellular transfer was not provene Accessed by the measurement of calcium Xuxesf Proposed from the observation of a seamless transition between microtubules of the bridge with microtubular networks of both connected cells
ND: not determined
Cell type TNT-like structures
Cytoskeletal components
Membrane continuity/“open-ended”
Cargo
PC12 cells (Rustom et al. 2004) F-actin, myosin Va (+)b Endosome-related organelles, lipid-anchored proteins (EGFP-f), EGFP-actin
Jurkat T cells (Sowinski et al. 2008) F-actin (¡)b,e HIV-1 protein Gag
123
542 Histochem Cell Biol (2008) 129:539–550
ended” cellular nanotubes (Fig. 3b). Other identiWed close-ended nanotubular bridges that share some morphologicalfeatures with TNT are murine leukemia virus-induced“Wlopodial bridges” (Sherer et al. 2007) (Table 1).
Interestingly, it has been shown that diVerent classes ofnanotubes exist even within a single cell type. In the case ofmacrophages, not only thin, F-actin-containing bridgessimilar to other TNT-like structures were detected, but alsoa thicker type of cellular connection (¸0.7 �m diameter),which contained both F-actin and microtubules (Önfeltet al. 2006). This thicker type of connection was suggestedto be open ended due to a seamless transition of the micro-tubules inside the bridge with the microtubular networks ofboth connected cells (Önfelt et al. 2006) (Table 1). Thin,and thicker microtubule-containing connections were alsofound between prostate cancer cells (Vidulescu et al. 2004).
Finally, another type of long and thin F-actin-containingcellular protrusions that share striking features with TNT-like intercellular bridges, are cytonemes. These membranenanotubes were discovered in Drosophila imaginal wingdiscs, emanating from the periphery of the columnar cellsheet toward the signaling center associated with the ante-rior/posterior border (Ramírez-Weber and Kornberg 1999).They are thought to extend toward the target cells by che-motaxis (Ramírez-Weber and Kornberg 1999) (Fig. 3c) toaccomplish the receptor-mediated uptake of the secretedmorphogen Decapentaphlegic (Dpp) during spatial pattern-
ing (Hsiung et al. 2005). Once endocytosed, the morphogenis thought to be delivered to the cell body by retrogradetransport along the cytoneme (Fig. 3c1). Thus, these struc-tures are proposed to fulWll an important task in long-rangecell-to-cell signaling during embryonic development. Incontrast to TNT-like structures, cytonemes have not beenshown to physically bridge cells. Nevertheless, functionallysimilar structures, emanating from tracheal cells inresponse to Branchless (Bnl)-Fibroblast Growth Factor(FGF) signaling, appear to do so (Sato and Kornberg 2002).
The identiWcation of TNT-like structures in cell cultureshas been mainly based on the morphological criteria deW-ned for PC12 cells (Rustom et al. 2004). In some cases, aproper comparison was hindered by their limited character-ization. Nonetheless, heterogeneity regarding formation,structure, and functional properties across cell types andeven within a given cell type has emerged (Table 1). Cer-tainly, more detailed information is necessary for a properclassiWcation of all these structures and thus, as things arenow, we shall refer to them as TNT-like structures, irre-spective of whether membrane continuity was observed.
Formation of TNT-like structures
TNT-like structures form de novo between cells on a time-scale of several minutes by apparently two distinct mecha-
Fig. 2 Transmission electron micrographs (TEM) showing the ultra-structure of distinct TNT-like bridges in diVerent cell types. (a) Open-ended TNT connecting two PC12 cells reconstructed from images oftwo consecutive 80 nm sections. The boxed areas are shown as highermagniWcation images (a1, a2). A continuous membrane is observed be-tween the nanotube and the plasma membrane of the two connectedcells. ModiWed from Rustom et al. (2004) Science 303:1007–1010. (b)
Close-ended TNT-like bridge connecting two T cells and displaying ajunctional border, reconstructed from images of 13 consecutive 60 nmsections. The boxed areas are shown as higher magniWcation images(b1, b2). The nanotube formed by one cell (b2) protrudes into an invag-ination (arrowhead) of the connected cell (b1). ModiWed from Sowin-ski et al. (2008) Nat Cell Biol 10:211–219. Scale bars, a, b, 1 �m; a1,a2, b1, b2, 500 nm
123
Histochem Cell Biol (2008) 129:539–550 543
nisms. With respect to the Wrst mechanism, initial studieson PC12 cells showed that TNT could be established by aseemingly directed outgrowth of swaying Wlopodia-likeprotrusion(s) toward a neighboring cell. Once contact ismade, a single dilated and bended bridge is often observedalong with the degeneration of remainder protrusions. Thisbended structure is then remodeled into a straight and thinbridge showing the characteristic morphology of typicalTNT (Rustom et al. 2004). One can speculate that theseemingly directed Wlopodia outgrowth preceding TNT for-mation is under control of chemotactic guidance. Evidencefor this is provided for related processes like the formationof cytonemes toward a Bnl-FGF gradient in in vitro cul-tures of Drosophila cells (Ramírez-Weber and Kornberg1999) (Fig. 3c) or murine leukemia virus (MLV)-inducednanotubular bridges emanating from non-infected Wbroblastcells toward infected cells (Sherer et al. 2007). The obser-vation that TNT-like bridges also emerge when attachedPC12 (Rustom et al. 2004) or immune cells (Önfelt et al.2004, 2006; Sowinski et al. 2008) depart from each other,
led to the proposal of a second mechanism of TNT forma-tion (Gerdes et al. 2007; Önfelt et al. 2004), which a prioriexcludes any dependence on chemotactic guidance of Wlo-podia. Subsequent detailed studies on T cells revealed thatcell-to-cell interaction for at least a few minutes before dis-lodging is required for successful nanotubular bridge for-mation (Sowinski et al. 2008). It is of note that, in contrastto PC12 cells, a Wlopodium-dependent mechanism of nano-tubular bridge formation between immune cells was notreported to date (Önfelt et al. 2004, 2006; Sowinski et al.2008).
Actin polymerization is thought to be a key event forboth mechanisms of TNT formation (Gerdes et al. 2007). Itdrives Wlopodia outgrowth important in the Wrst mechanismand is most probably important for the stabilization ofTNT-like structures emerging by the second mechanism. Insupport of this, TNT formation is not observed in the pres-ence of actin-depolymerizing drugs (Rustom et al. 2004),and TNT-like structures between T cells only form if thecells diverge with a speed below that of processes driven by
Fig. 3 Schematic representa-tions of three distinct nanoscaled cellular protrusions and pro-posed modes of cell-to-cell com-munication. (a) A TNT-mediating membrane continuity between cells. (a1) Organelles like endocytic vesicles and mito-chondria are transported uni-directionally between cells by an actin-dependent mechanism. (b) Nanotubular bridge between cells displaying a junctional bor-der. (b1) Distinct viral particles are transported either at the sur-face of the nanotube by a recep-tor-dependent mechanism using actin retrograde Xow or inside the cellular nanotube by an ac-tin-dependent mechanism. (c) Cellular nanotube (cytoneme) extending toward a target cell by chemotaxis. (c1) Signaling mole-cules secreted by the target cell are proposed to be endocytosed by a receptor-mediated mecha-nism at the tip of the cytoneme and transported in a retrograde manner toward the cell body of the receiving cell. The arrows (a–c) indicate the direction of transfer
123
544 Histochem Cell Biol (2008) 129:539–550
actin polymerization (Cameron et al. 1999; Sowinski et al.2008). Zhu et al. (2005) obtained further evidence that actinpolymerization and TNT formation are linked by demon-strating their concurrent induction by H2O2 in primary cul-tures of rat astrocytes.
TNT-like structures have a dynamic nature leading toonly transient bridges as observed in in vitro cultures. Thenanotubular bridges between T cells (Sowinski et al. 2008),PC12 cells or normal rat kidney (NRK) cells (Bukoresht-liev, N., Gerdes, H.-H., unpublished data) have variablelifetimes, ranging from a few minutes to less than 60 minfor the former and even up to several hours for the lattertwo. These considerable diVerences in lifetime, even at thecellular level, may reXect the existence of diVerent sub-classes of TNT-like structures. It is interesting to speculatethat the fusion of the nanotube with the plasma membraneof the connected cell is part of a time-dependent maturationstep leading to such heterogeneity.
Studies on model artiWcial membrane tubes such as thosecreated by pulling tethers from synthetic lipid vesicles orcellular plasma membrane provide a complementaryapproach to obtain mechanistic insights into the formationof TNT-like structures. The morphology of such nanotubesresembles that of TNT-like structures and it seems plausi-ble that the same physical laws govern the formation andarchitecture of both structures. The exploitation of suchin vitro models revealed that the elongation of plasmamembrane nanotubes requires a membrane Xow from thecell plasma membrane into the growing tube. It is sug-gested that cells maintain a membrane reservoir (e.g.ruZes, invaginations), controlled by the cytoskeleton, toprovide a buVer against membrane tension over severalmicrometers of tube elongation (Raucher and Sheetz 1999;Sun et al. 2005). Thus, the available membrane reservoirmay restrict the number, total length, and lifetime of TNT-like connections for a given cell.
TNT-like structures are conduits for the delivery of cargo
An obvious advantage of establishing direct bridgesbetween cells is an improvement in selective communica-tion even over long distances. In agreement with thisassumption, the common feature of all TNT-like structurescharacterized in more detail, is the transfer of cargobetween connected cells (Table 1).
Cellular components
Direct evidence for the intercellular exchange of cargo viaTNT-like structures was obtained by video-microscopicstudies. By employing Xuorescent dyes, organelles belong-
ing to the endosomal/lysosomal system (Rustom et al.2004) (Gurke, S., Gerdes, H.-H., unpublished data) as wellas mitochondria (Koyanagi et al. 2005) were shown totraYc uni-directionally along TNT-like structures betweencells over long distances (Fig. 3a1) (Table 1). The occur-rence of an intercellular transfer of organelles raises thequestion as to what kind of information is transferred andhow this information is integrated in the target cell. In thecase of the endosomal system, early endocytic vesicles areparticularly interesting candidates for TNT-mediated deliv-ery. These organelles are one of the major reloading pointsfor a variety of signaling complexes resulting from theendocytosis of activated cell surface receptors (Miaczynskaet al. 2004). Once delivered to the target cells, the trans-ferred endocytic vesicles are able to fuse with their counter-parts (Rustom et al. 2004), and provide a potential way tointegrate information. In essence, the transfer of endocyticorganelles propagates signaling information from the singlecell to a larger community, which may lead to coordinatedcell behavior. Mitochondria were shown to transfer viaTNT from neonatal rat CM to adult human EPC and it wassuggested that this event could contribute to the acquisitionof a cardiomyogenic phenotype by the progenitor cells(Koyanagi et al. 2005) (Fig. 4). Moreover, the active trans-fer of this organelle from adult stem cells and somatic cellsto mammalian cells with non-functional mitochondria wasshown to rescue aerobic respiration in the latter (Csordás2006; Spees et al. 2006). The obtained data are consistentwith a TNT-related delivery, although alternative mecha-nisms were not excluded.
In addition to the transfer of organelles, plasma mem-brane components such as lipid-anchored proteins can lat-erally transfer along the TNT-like bridge into the plasma
Fig. 4 Emerging physiological implications of TNT-like structures
123
Histochem Cell Biol (2008) 129:539–550 545
membrane of connected cells. This was demonstrated forboth the inner and outer leaXets of the plasma membrane byusing glycosylphosphatidylinositol conjugated to GFP(GPI)-GFP (Önfelt and Davis 2004; Önfelt et al. 2004) andthe farnesylation signal of c-Ha-Ras fused to the C-termi-nus of EGFP (EGFP-f) (Rustom et al. 2004), respectively(Table 1). Such a transport is consistent with membranecontinuity between connected cells. Surface receptors couldalso be transferred directly by lateral diVusion in the planeof the cell surface, in addition to an endosome-dependentdelivery. Interestingly, the transfer of membrane proteinsbetween myeloid cells was found to involve scavengerreceptors (Dr. R. D. Salter, Univ. of Pittsburgh, personalcommunication). It is of note that major histocompatibilitycomplex (MHC) class I receptors were detected at theentire length of TNT-like bridges connecting immune cells(Önfelt et al. 2004; Watkins and Salter 2005). This pointsto a key role of TNT in a faster and more eYcient presenta-tion of antigens, in particular at the immunological synapse(Groothuis et al. 2005; Williams et al. 2007) (Fig. 4), deW-ned as a highly specialized interface between immune cells(Norcross 1984; Paul and Seder 1994). In this case, the lat-eral transfer of only a few molecules can fulWll importantroles in a coordinated immune response. Interestingly, inanalogy to TNT, membrane continuity was found at theimmunological synapse (Stinchcombe et al. 2001). Thecharacterized “membrane bridges” had a diameter of50–95 nm as observed by electron microscopy (Stinchcombeet al. 2001).
Finally, the proposed model of membrane continuitysuggests a cytoplasmic bridge in analogy to PD, which per-mits the free Xow of soluble molecules up to a certain sizelimit deWned by the free space along the interior of thenanotube. In support of this view, EGFP-actin was trans-ferred between TNT-connected PC12 cells (Rustom et al.2004), GFP between CM and EPC (Koyanagi et al. 2005),and the Xuid phase marker Lucifer yellow between THP-1monocytes (Watkins and Salter 2005) (Table 1). Notably,both GFP and the small dye molecule calcein, with amolecular weight of 400 Da, were apparently impeded todiVuse passively between TNT-connected PC12 cells(Rustom et al. 2004). Thus, TNT-like structures, like PD,may have a variable size exclusion limit depending on thecell type (Gerdes et al. 2007). With regard to soluble mark-ers, certainly the most physiologically relevant Wnding wasthe TNT-dependent propagation of calcium signalsbetween THP-1 monocytes as well as from DC to THP-1monocytes (Watkins and Salter 2005) (Table 1). Stimula-tion by contact or exogenous soluble stimulators of bothDC and THP-1 monocytes led to a calcium Xux that propa-gated through a network of TNT-connected cells up to500 �m away from the point of stimulation. The elicitedcalcium wave propagated uni-directional with an initial
speed of 35 �m/s that slowed down rapidly to 10–15 �m/s(Watkins and Salter 2005). This rather low and rapidlydecreasing speed supports a TNT-mediated signal spreadalong the cellular network, which is driven by a chemicalgradient rather than an action potential. Such a view alsocorroborates a TNT-mediated membrane continuity provi-ding a cytoplasmic bridge between connected cells(Table 1). Most remarkably, the transfer of calcium ionsthrough TNT led to the Xattening and membrane extensionof the receiving cells, as occurs during physiologicalresponses in preparation for phagocytosis (Watkins andSalter 2005). Thus, this study provided the Wrst evidencefor functional connectivity accomplished by TNT and sug-gests a crucial role of these structures in immune defense(Fig. 4).
Pathogens
An emerging topic is the eYcient spread of pathogens, likebacteria and mammalian viruses, along cell membrane pro-trusions (Figs. 3, 4). By using established or induced nano-tubular connections between cells, pathogens circumventthe rate-limiting step of diVusion and eYciently acceleratetheir spreading (Sherer et al. 2007; Sowinski et al. 2008).Among the viruses shown to exploit cellular protrusions totransfer from cell-to-cell are the vaccinia virus (Cudmoreet al. 1995,. 1996), pseudorabies virus (Favoreel et al.2005, 2006), the herpes simplex virus (La Boissière et al.2004), MLV (Sherer et al. 2007) and human immunodeW-ciency virus type 1 (HIV-1) (Sowinski et al. 2008). In thefollowing, we will focus on the latter two, which were char-acterized in much detail and gave rise to two diVerentmechanisms of virus transmission.
Sowinski et al. (2008) proposed membrane nanotubesbetween T cells as a new route for HIV-1 transmission(Fig. 3b1), in addition to cellular contact points known asvirological synapses (Jolly et al. 2004; Jolly and Sattentau2004; Sol-Foulon et al. 2007). Using a recombinant HIV-1expressing Gag-GFP, they demonstrated its uni-directionalmove through innate nanotubular bridges and subsequenttransmission into connected T cells (Sowinski et al. 2008)(Table 1). Furthermore, the requirement of cell surfacereceptors for viral transmission implied, in agreement withultra-structural data, a junctional border between the nano-tube and the connected T cell. It should be emphasized thatthe transmission of HIV-1 through nanotubular bridges canminimize the exposure of the virus to extracellular antibodiesor complement. However, a nanotube-dependent spread mayopen new avenues in the development of antiviral drugs.
A diVerent, even though nanotube-dependent, mecha-nism of transmission was reported for MLV (Sherer et al.2007). The cell-to-cell transmission of MLV was directlyvisualized by infecting cultured cells with a recombinant
123
546 Histochem Cell Biol (2008) 129:539–550
virus expressing the viral capsid protein Gag and the enve-lope protein Env as a fusion construct with Xuorescent pro-teins. MLV induces the outgrowth of thin and longWlopodia from uninfected cells toward infected cells(Sherer et al. 2007). These Wlopodia are proposed to stablyanchor to the cell body of an infected cell by viral envelopeglycoprotein (Env)/receptor interactions. Subsequently,MLV moves along the outer surface of these nanotubularbridges to reach the cell body of target cells, where it Wnallyenters (Sherer et al. 2007) (Fig. 3b1). In addition to viruses,bacteria were reported to surf along TNT-like structuresconnecting macrophages (Önfelt et al. 2006) (Table 1).
Mechanism of transfer
It is possible that more than one transfer mechanism forshipping cargo via TNT-like bridges has evolved consider-ing their structural diversity and the multitude oftransferred cargo. For organelles and viruses, an F-actin-dependent, uni-directional transport prevails (Fig. 3), withthe exception of a microtubule-dependent, bi-directionalmovement of organelles in thick bridges connecting macro-phages (Önfelt et al. 2006). In the case of endocytic organ-elles, an acto-myosin-dependent transport system ispresumed (Fig. 3a1) due to the presence of the barbed-enddirected actin-speciWc motor myosin Va in TNT (Rustomet al. 2004; Zhu et al. 2005) (Bukoreshtliev, N., Gerdes,H.-H., unpublished data) (Table 1) partially co-localizing withthe respective organelles (Rustom et al. 2004). This modelis supported by the low transport velocity in the range ofactin-dependent transport (Rustom et al. 2004) and impliesthat the F-actin Wbers inside the bridge have the same polar-ity. Also HIV-1 particles were suggested to move in anactin-dependent manner through nanotubular bridges(Sowinski et al. 2008) (Fig. 3b1). Interestingly, even in thecase of MLV surWng along Wlopodia bridges with a speed»7 times slower as compared to that of HIV-1 within thebridges, an acto-myosin transfer mechanism appears to bein place (Sherer et al. 2007). This results in the uni-direc-tional movement of all viral particles from infected to targetcells, presumably using myosin II controlled retrogradeXow of actin and cognate Env-receptor interactions(Fig. 3b1). In contrast to organelles and viruses, the uni-directional propagation of calcium waves is likely to bedriven by a chemical gradient rather than an active trans-port mechanism. Passive diVusion is likely to control thetransfer of other small cytoplasmic molecules and alsosome plasma membrane components, if membrane conti-nuity is present.
Regarding the modality by which shipped cargo entersthe target cell, two diVerent scenarios are conceivable. Inthe case of membrane continuity as observed in PC12 cells,
an open-ended transport along cytoskeleton elements with-out border crossing is envisaged, consistent with theobserved uniform movement of transferred organelles(Rustom et al. 2004). In analogy to PD, gating mechanismsmay exist to control this open-ended transport (Gerdes et al.2007). In the case of close-ended cell-to-cell bridges con-taining junctional borders, as are the innate nanotubes con-necting T cells and the MLV-induced cellular bridges, thecargo has to traverse the plasma membrane boundary. Thisis consistent with the Wnding that a receptor-dependenttransmission of HIV-1 Gag could be blocked by the HIV-1fusion inhibitor T20 (Sowinski et al. 2008).
Ultra-Wne cellular extensions in tissue
Does intercellular communication mediated by TNT-likestructures play a physiological role in tissue? This is proba-bly the most interesting and important question that needsto be addressed in the near future. The absence of a TNT-speciWc marker does not allow an explicit answer to thisquestion yet. However, studies employing cell culturesalready point to important physiological implications ofTNT-like structures in intercellular communication(reviewed in Gerdes et al. (2007)) (Fig. 4). It could beargued that cell culture models are rough representations ofphysiological systems and that TNT-like structures are thesole result of stress conditions in this in vitro situation.However, the widely documented richness of cellular pro-trusions in tissue implicated in intercellular communication(Rørth 2003) (Table 2) disagrees with such a possibilityand instead supports an in vivo function. Cellular exten-sions have been documented in diverse tissues and duringvarious physiological processes such as embryogenesis,stem cell diVerentiation, cell migration, and wound healing.
The formation of cellular extensions (“pseudopods”)during embryogenesis became Wrst evident by studying seaurchin morphogenesis (Gustafson and Wolpert 1967;Miller et al. 1995), suggesting a function of these structuresin force generation or cell recognition. During subsequentstudies over the last decades, a variety of similar cellularextensions were found to be associated with developmentalprocesses. Interestingly, most of them appear to be actin-rich as TNT-like structures. The Drosophila wing imaginaldisc provides an attractive system for studying cell protru-sions within a developing tissue. The Wrst described exam-ple of such protrusions were the F-actin-containingcytonemes extending from the periphery of the columnarcell sheet toward the signaling center within this layer(Ramírez-Weber and Kornberg 1999). This discoveryinspired further studies describing other cellular protrusionsin the wing disc extending not only within (Chou and Chien2002; Demontis and Dahmann 2007) but also between cell
123
Histochem Cell Biol (2008) 129:539–550 547
Tab
le2
Sele
cted
exa
mpl
es o
f ce
ll p
rotr
usio
ns in
tiss
ue
aD
ata
from
Mill
er e
tal.
(199
5)b
Dat
a fr
om J
acin
to e
tal.
(200
0)c
The
se m
olec
ules
wer
e sh
own
to lo
caliz
e in
cel
l pro
trus
ions
, but
thei
r in
terc
ellu
lar
tran
sfer
was
not
pro
ven
dD
ata
from
Mis
geld
eta
l. (2
002)
ND
: not
det
erm
ined
Pro
trus
ions
T
issu
e C
ytos
kele
tal
com
pone
nts
Car
go
Len
gth
[�m
] D
iam
eter
[n
m]
Gro
wth
spe
ed
[�m
/min
]
Thi
n, e
long
ated
, ac
tive W
lopo
dia
and
lam
ellip
odia
Pri
mar
y m
esen
chym
e se
a ur
chin
em
bryo
s (G
usta
fson
and
Wol
pert
196
7; M
iller
eta
l. 19
95)
F-ac
tina
ND
80
a20
0–40
0b 25
a
Cyt
onem
es
Dro
soph
ila
win
g im
agin
al d
iscs
(R
amír
ez-W
eber
and
Kor
nber
g 19
99)
F-ac
tin
ND
<
700
200
15
Cyt
onem
e-lik
e M
ouse
lim
b bu
d ce
lls
(Ram
írez
-Web
er a
nd K
ornb
erg
1999
);
Dro
soph
ila
win
g im
agin
al d
iscs
(Ja
cint
o et
al. 2
000)
F-
acti
nbN
D
10b
ND
1b
Api
cal p
erip
odia
l ex
tens
ions
D
roso
phil
a w
ing/
eye
imag
inal
dis
cs
(Gib
son
and
Sch
ubig
er 2
000)
M
icro
tubu
les
Mito
chon
dria
c 5–
30
ND
N
D
Myo
podi
a D
roso
phil
a em
bryo
nic
mus
cle
cells
(R
itzen
thal
er e
tal.
2000
; Rit
zent
hale
r an
d C
hiba
200
3)
F-ac
tin
ND
40
N
D
ND
Myo
podi
a-li
ke
Mou
se tr
ansv
erse
mus
cle
of th
orax
(M
isge
ld e
tal.
2002
);
Xen
opus
(K
ullb
erg
etal
. 197
7)
ND
N
D
10d
ND
N
D
Cel
lula
r ex
tens
ions
D
roso
phil
a ey
e im
agin
al d
iscs
(C
hou
and
Chi
en 2
002)
F-
acti
n Sc
abro
us v
esic
les
40
ND
N
D
Api
cal a
nd la
tera
l ce
ll pr
otru
sion
D
roso
phil
a w
ing
imag
inal
dis
cs
(Dem
onti
s an
d D
ahm
ann
2007
) F-
acti
n G
FP-R
ab5c , G
FP-
Tkv
c1–
10
200
ND
123
548 Histochem Cell Biol (2008) 129:539–550
layers (Demontis and Dahmann 2007; Gibson and Schubi-ger 2000). In a very recent study on live wing discs, Demon-tis and Dahmann (2007) observed two new types ofcellular extensions, i.e. lateral protrusions interconnectingdistant cells of the columnar epithelium, and protrusions ofthe apical surface connecting to the squamous epithelium.Furthermore, ectopically expressed GFP-actin and the earlyendosome marker GFP-Rab 5 were detected inside theseprotrusions, resembling the Wnding made for TNT-likestructures between cultured cells. These diVerent protru-sions may be involved in the intercellular signaling neces-sary for growth and patterning of wing imaginal discs(Demontis and Dahmann 2007).
Thin, actin-based membrane extensions were also foundto be crucial for embryonic dorsal closure in Drosophila(Jacinto et al. 2000). Data obtained by confocal imaging ofliving Xy embryos suggest that these long protrusions par-ticipate in both the mechanics of epithelial adhesion and inthe search for a correct partner within the opposing epithe-lium. Likewise, related fundamental processes such as gas-trulation, neural crest closure (Bard 1992), and woundhealing (Wood et al. 2002) seem to involve similar actin-rich membrane extensions.
Other examples of actin-based cellular extensions duringdevelopment are the so-called myopodia in Drosophilaembryonic muscle cells that can be observed in wholemount. These dynamic postsynaptic microprocesses, withlengths up to 40 �m and highly sensitive to photoillumina-tion, cluster at the site of motoneuron innervation whileinteracting with presynaptic Wlopodia. It was proposed thatthey play a role in the process of synaptic target recognitionby contributing to direct long-distance cellular communica-tion (Ritzenthaler et al. 2000; Ritzenthaler and Chiba 2003).Importantly, similar structures have been described in Xeno-pus (Kullberg et al. 1977) and mouse (Misgeld et al. 2002).
Finally, long, thin cellular extensions are also associatedwith guided cellular migration during development (Ribe-iro et al. 2002; Sato and Kornberg 2002), immune defensesuch as the migration of macrophages through the endothe-lial cell layer (Blue and Weiss 1981) and tumor cell inva-sion (Vignjevic and Montagnac 2008). Certainly, there aremany more examples of long cellular nanotubes in tissueinvolved in cellular communication besides the ones listedhere (Table 2). This suggests that extensions in tissue are arather common feature and that cells in culture preserve theability to form similar, physiologically relevant structures.But, an important question remains: Do TNT-like struc-tures, observed in cell culture models, have mechanisticallyand/or functionally related counterparts in live tissue? Fur-thermore, with respect to TNT, which have been found tomediate membrane continuity between cultured cells, it willbe of foremost interest to Wnd out, if this type of connectionalso exists in vivo.
Acknowledgments We thank Simone Reber for her contribution tothe manuscript and all group members for stimulating discussions. Fur-thermore, we are grateful to Dr. Salter, University of Pittsburgh, forproviding unpublished data. This work is being supported by the Nor-wegian Research Council (HHG), the Norwegian Cancer Society(JFVB, HHG), the University of Bergen, Norway (HHG), and theGottlieb Daimler- and Karl Benz-Foundation, Germany (SG).
References
Ambudkar SV, Sauna ZE, Gottesman MM, Szakacs G (2005) A novelway to spread drug resistance in tumor cells: functional intercel-lular transfer of P-glycoprotein (ABCB1). Trends Pharmacol Sci26:385–387
Balunka F, Hlavacka A, Volkmann D, Menzel D (2004a) Getting con-nected: actin-based cell-to-cell channels in plants and animals.Trends Cell Biol 14:404–408
Balunka F, Volkmann D, Barlow PW (2004b) Cell bodies in a cage.Nature 428:371
Balunka F, Volkmann D, Barlow PW (2004c) Eukaryotic cells andtheir Cell Bodies: Cell Theory revised. Ann Bot (Lond) 94:9–32
Bard J (1992) Morphogenesis: the cellular and molecular processes ofdevelopmental anatomy. Cambridge Univ Press, Cambridge, U.K
Bjerkvig R, Tysnes BB, Aboody KS, Najbauer J, Terzis AJ (2005)Opinion: The origin of the cancer stem cell: current controversiesand new insights. Nat Rev Cancer 5:899–904
Blue J, Weiss L (1981) Electron microscopy of the red pulp of the dogspleen including vascular arrangements, periarterial macrophagesheaths (ellipsoids), and the contractile, innervated reticularmeshwork. Am J Anat 161:189–218
Cameron LA, Footer MJ, van Oudenaarden A, Theriot JA (1999)Motility of ActA protein-coated microspheres driven by actinpolymerization. Proc Natl Acad Sci USA 96:4908–4913
Cilia ML, Jackson D (2004) Plasmodesmata form and function. CurrOpin Cell Biol 16:500–506
Csordás A (2006) Mitochondrial transfer between eukaryotic animalcells and its physiologic role. Rejuvenation Res 9:450–454
Cudmore S, Cossart P, GriYths G, Way M (1995) Actin-based motilityof vaccinia virus. Nature 378:636–638
Cudmore S, Reckmann I, GriYths G, Way M (1996) Vaccinia virus: amodel system for actin-membrane interactions. J Cell Sci109:1739–1747
Demontis F (2004) Nanotubes make big science. PLoS Biol 2:E215Demontis F, Dahmann C (2007) Apical and lateral cell protrusions
interconnect epithelial cells in live Drosophila wing imaginaldiscs. Dev Dyn 236:3408–3418
Favoreel HW, Van Minnebruggen G, Adriaensen D, Nauwynck HJ(2005) Cytoskeletal rearrangements and cell extensions inducedby the US3 kinase of an alphaherpesvirus are associated with en-hanced spread. Proc Natl Acad Sci USA 102:8990–8995
Favoreel HW, Van Minnebruggen G, Van de Walle GR, Ficinska J,Nauwynck HJ (2006) Herpesvirus interference with virus-speciWcantibodies: bridging antibodies, internalizing antibodies, and hid-ing from antibodies. Vet Microbiol 113:257–263
Freund D, Bauer N, Boxberger S, Feldmann S, Streller U, Ehninger G,Werner C, Bornhäuser M, Oswald J, Corbeil D (2006) Polariza-tion of human hematopoietic progenitors during contact with mul-tipotent mesenchymal stromal cells: eVects on proliferation andclonogenicity. Stem Cells Dev 15:815–829
Gallagher KL, Benfey PN (2005) Not just another hole in the wall:understanding intercellular protein traYcking. Genes Dev19:189–195
123
Histochem Cell Biol (2008) 129:539–550 549
Gerdes H-H, Bukoreshtliev NV, Barroso JF (2007) Tunneling nano-tubes: a new route for the exchange of components between ani-mal cells. FEBS Lett 581:2194–2201
Gibson MC, Schubiger G (2000) Peripodial cells regulate proliferationand patterning of Drosophila imaginal discs. Cell 103:343–350
Groothuis TA, Griekspoor AC, Neijssen JJ, Herberts CA, Neefjes JJ(2005) MHC class I alleles and their exploration of the antigen-processing machinery. Immunol Rev 207:60–76
Guo G-Q, Zheng G-C (2004) Hypotheses for the functions of intercel-lular bridges in male germ cell development and its cellular mech-anisms. J Theor Biol 229:139–146
Gustafson T, Wolpert L (1967) Cellular movement and contact in seaurchin morphogenesis. Biol Rev Camb Philos Soc 42:442–498
Hodneland E, Lundervold A, Gurke S, Tai X-C, Rustom A, Gerdes H-H (2006) Automated detection of tunneling nanotubes in 3D im-ages. Cytometry A 69:961–972
Jacinto A, Wood W, Balayo T, Turmaine M, Martinez-Arias A, MartinP (2000) Dynamic actin-based epithelial adhesion and cell match-ing during Drosophila dorsal closure. Curr Biol 10:1420–1426
Jolly C, Sattentau QJ (2004) Retroviral spread by induction of virolog-ical synapses. TraYc 5:643–650
Jolly C, KasheW K, Hollinshead M, Sattentau QJ (2004) HIV-1 cell tocell transfer across an Env-induced, actin-dependent synapse.J Exp Med 199:283–293
Kajstura J, Leri A, Bolli R, Anversa P (2006) Endothelial progenitorcells: neovascularization or more? J Mol Cell Cardiol 40:1–8
Koyanagi M, Brandes RP, Haendeler J, Zeiher AM, Dimmeler S(2005) Cell-to-cell connection of endothelial progenitor cells withcardiac myocytes by nanotubes: a novel mechanism for cell fatechanges? Circ Res 96:1039–1041
Kullberg RW, Lentz TL, Cohen MW (1977) Development of the my-otomal neuromuscular junction in Xenopus laevis: an electro-physiological and Wne-structural study. Dev Biol 60:101–129
La Boissière S, Izeta A, Malcomber S, O’Hare P (2004) Compartmen-talization of VP16 in cells infected with recombinant herpes sim-plex virus expressing VP16-green Xuorescent protein fusionproteins. J Virol 78:8002–8014
Levchenko A, Mehta BM, Niu X, Kang G, Villafania L, Way D, Poly-carpe D, Sadelain M, Larson SM (2005) Intercellular transfer ofP-glycoprotein mediates acquired multidrug resistance in tumorcells. Proc Natl Acad Sci USA 102:1933–1938
Magalhães AC, Baron GS, Lee KS, Steele-Mortimer O, Dorward D,Prado MAM, Caughey B (2005) Uptake and neuritic transport ofscrapie prion protein coincident with infection of neuronal cells.J Neurosci 25:5207–5216
Miaczynska M, Pelkmans L, Zerial M (2004) Not just a sink: endo-somes in control of signal transduction. Curr Opin Cell Biol16:400–406
Miller J, Fraser SE, McClay D (1995) Dynamics of thin Wlopodia dur-ing sea urchin gastrulation. Development 121:2501–2511
Misgeld T, Burgess RW, Lewis RM, Cunningham JM, Lichtman JW,Sanes JR (2002) Roles of neurotransmitter in synapse formation:development of neuromuscular junctions lacking choline acetyl-transferase. Neuron 36:635–648
Norcross MA (1984) A synaptic basis for T-lymphocyte activation.Ann Immunol (Paris) 135D:113–134
Önfelt B, Davis DM (2004) Can membrane nanotubes facilitate com-munication between immune cells? Biochem Soc Trans 32:676–678
Önfelt B, Nedvetzki S, Benninger RK, Purbhoo MA, Sowinski S,Hume AN, Seabra MC, Neil MA, French PM, Davis DM (2006)Structurally distinct membrane nanotubes between human macro-phages support long-distance vesicular traYc or surWng of bacte-ria. J Immunol 177:8476–8483
Paul WE, Seder RA (1994) Lymphocyte responses and cytokines. Cell76:241–251
Ramírez-Weber F-A, Kornberg TB (1999) Cytonemes: cellular pro-cesses that project to the principal signaling center in Drosophilaimaginal discs. Cell 97:599–607
Ratajczak MZ, Kucia M, Dobrowolska H, Wanzeck J, Reca R, Ratajc-zak J (2006) Emerging concept of cancer as a stem cell disorder.CEJB 1:73–87
Raucher D, Sheetz MP (1999) Characteristics of a membrane reservoirbuVering membrane tension. Biophys J 77:1992–2002
Ribeiro C, Ebner A, AVolter M (2002) In vivo imaging reveals diVer-ent cellular functions for FGF and Dpp signaling in trachealbranching morphogenesis. Dev Cell 2:677–683
Ritzenthaler S, Chiba A (2003) Myopodia (postsynaptic Wlopodia) par-ticipate in synaptic target recognition. J Neurobiol 55:31–40
Ritzenthaler S, Suzuki E, Chiba A (2000) Postsynaptic Wlopodia inmuscle cells interact with innervating motoneuron axons. NatNeurosci 3:1012–1017
Rørth P (2003) Communication by touch: role of cellular extensions incomplex animals. Cell 112:595–598
Ruiz-Medrano R, Xoconostle-Cazares B, Kragler F (2004) The plas-modesmatal transport pathway for homeotic proteins, silencingsignals and viruses. Curr Opin Plant Biol 7:641–650
Rustom A, SaVrich R, Markovic I, Walther P, Gerdes H-H (2004)Nanotubular highways for intercellular organelle transport. Sci-ence 303:1007–1010
Sato M, Kornberg TB (2002) FGF is an essential mitogen and chemo-attractant for the air sacs of the Drosophila tracheal system. DevCell 3:195–207
Sol-Foulon N, Sourisseau M, Porrot F, Thoulouze M-I, Trouillet C,Nobile C, Blanchet F, di Bartolo V, Noraz N, Taylor N, AlcoverA, Hivroz C, Schwartz O (2007) ZAP-70 kinase regulates HIVcell-to-cell spread and virological synapse formation. Embo J26:516–526
Sowinski S, Jolly C, Berninghausen O, Purbhoo MA, Chauveau A,Köhler K, Oddos S, Eissmann P, Brodsky FM, Hopkins C, ÖnfeltB, Sattentau Q, Davis DM (2008) Membrane nanotubes physi-cally connect T cells over long distances presenting a novel routefor HIV-1 transmission. Nat Cell Biol 10:211–219
Spees JL, Olson SD, Whitney MJ, Prockop DJ (2006) Mitochondrialtransfer between cells can rescue aerobic respiration. Proc NatlAcad Sci USA 103:1283–1288
Stinchcombe JC, Bossi G, Booth S, GriYths GM (2001) The immuno-logical synapse of CTL contains a secretory domain and mem-brane bridges. Immunity 15:751–761
Sun M, Graham JS, Hegedüs B, Marga F, Zhang Y, Forgacs G, Grand-bois M (2005) Multiple membrane tethers probed by atomic forcemicroscopy. Biophys J 89:4320–4329
Tysnes BB, Bjerkvig R (2007) Cancer initiation and progression:involvement of stem cells and the microenvironment. BiochimBiophys Acta 1775:283–297
Vidulescu C, Clejan S, O’Connor KC (2004) Vesicle traYc throughintercellular bridges in DU 145 human prostate cancer cells. J CellMol Med 8:388–396
Vignjevic D, Montagnac G (2008) Reorganisation of the dendritic ac-tin network during cancer cell migration and invasion. SeminCancer Biol 18:12–22
123
550 Histochem Cell Biol (2008) 129:539–550
Watkins SC, Salter RD (2005) Functional connectivity between im-mune cells mediated by tunneling nanotubules. Immunity23:309–318
Williams GS, Collinson LM, Brzostek J, Eissmann P, Almeida CR,McCann FE, Burshtyn D, Davis DM (2007) Membranous struc-tures transfer cell surface proteins across NK cell immune syn-apses. TraYc 8:1190–1204
Wood W, Jacinto A, Grose R, Woolner S, Gale J, Wilson C, Martin P(2002) Wound healing recapitulates morphogenesis in Drosoph-ila embryos. Nat Cell Biol 4:907–912
Zhu D, Tan KS, Zhang X, Sun AY, Sun GY, Lee JC (2005) Hydrogenperoxide alters membrane and cytoskeleton properties and in-creases intercellular connections in astrocytes. J Cell Sci118:3695–3703
