sign in sign up
<<
Home , M1 protein, Viral neuraminidase, Vpr, Vpu protein

VIROLOGY 229, 1–11 (1997) ARTICLE NO. VY978451

MINIREVIEW

View metadata, citation and similar papers at core.ac.uk brought to you by CORE Do Vpu and of Human Immunodeficiency Type 1 and NB of B Virusprovided by Elsevier - Publisher Connector Have Activities in the Viral Life Cycles?

ROBERT A. LAMB*,†,1 and LAWRENCE H. PINTO‡

*Howard Hughes Medical Institute, †Department of Biochemistry, Molecular Biology, and Biology, and ‡Department of Neurobiology and Physiology, Northwestern University, Evanston, Illinois 60208-3500 Received December 19, 1996; returned to author for revision January 6, 1997; accepted January 13, 1997

In 1996, three viral proteins, NB of influenza B virus prior to molecular cloning of the cDNA and the expres- and Vpu and Vpr of human immunodeficiency virus type sion of the ion channel in heterologous systems. Thus, 1 (HIV-1), were reported in the literature as having an the characteristics of the channel properties could be ion channel activity (Ewart et al., 1996; Piller et al., 1996; anticipated. However, for candidate ion Schubert et al., 1996b; Sunstrom et al., 1996). Thus, in- channel activity, prior knowledge of their electrophysio-

cluding the well-characterized M2 pro- logical properties is rarely known. tein, there are now four viral proteins that are reported When a candidate viral ion channel protein is ex- to act as ion channels. However, demonstrating that a pressed in tissue culture cells or in oocytes of X. laevis viral protein has an ion channel activity that is relevant and a new cell surface conductance is recorded, it is to the life cycle of a virus or in causing viral pathogenesis usually assumed that this current results directly from is not a simple problem. expression of the viral protein. Although this may be a Ion channels are generally accepted to be integral correct conclusion, the caveat must be added that ex- membrane proteins that contain an aqueous accessible pression of several small integral membrane proteins of proteinaceous core (the channel pore) that can allow a both viral and cellular origin, at least in the oocyte sys- massive flux of ions (107 –108 ions per second) across a tem, is capable of regulating a channel(s) endogenous membrane. In contrast, some water-soluble peptides can to the host cell (Attali et al., 1993; Shimbo et al., 1995; promote the local breakdown of the membrane bilayer Tzounopoulos et al., 1995). Thus, unless a specific creating a lipidic pore with a surface formed completely blocker of the candidate channel is available, it may be or partially by polar lipid headgroups. The biophysical difficult to distinguish whether the current observed is characteristics of these lipidic pores are not as well de- intrinsic to the expressed viral protein. One approach fined as those of ion channels and are usually dependent toward overcoming this dilemma is to make mutations on the lipid composition of the membrane (Chernomordik in the presumptive channel pore region, seek changes et al., 1994). in properties of the channel (e.g., altered ionic selectivity, Ion channel activity is detected by measuring changes gating, or inactivity), and correlate these observations in cell surface currents, either of whole cells or of single with the effect of incorporating the same mutations into macromolecular complexes (single channel recordings). the virus genome. One of the simpler methods to analyze ion channel pro- To demonstrate that an ion channel activity is a bona teins is to express the proteins either in tissue culture fide property of the purified channel protein, attempts are cells or in oocytes of Xenopus laevis. This method re- often made to reconstitute the ion channel in planar lipid quires that the candidate protein is expressed at the bilayers. Thus, this technique can be used to confirm surface of cells in order to measure ion channel currents. properties of the ion channel activity measured by other For the ion channels of many specialized cells the nature means. If there is a concordance among the data, then of the surface currents had been measured in organs it can be safely concluded that the ion channel activity is intrinsic to the purified protein. However, in the ab- sence of supporting data interpretation of results ob- 1 To whom reprint requests should be addressed at Department of Biochemistry, Molecular Biology, and Cell Biology, Northwestern tained from planar bilayers must be treated cautiously, University, 2153 North Campus Dr., Evanston, IL 60208–3500. Fax: (847) as it is known that peptides capable of forming a-helical 491–2467. E-mail: [email protected]. bundles that do not form ion channels in mammalian

0042-6822/97 $25.00 1 Copyright ᭧ 1997 by Academic Press All rights of reproduction in any form reserved.

AID VY 8451 / 6a2a$$$241 02-13-97 14:46:20 vira AP: Virology 2 LAMB AND PINTO cells can induce ion channel activity when introduced transmembrane (TM) residues, and 54 C-terminal intra- into planar lipid bilayers (Tosteson et al., 1988, 1989). cellular cytoplasmic residues (Holsinger and Lamb, 1991; Furthermore, great care must be taken to ensure that Lamb et al., 1985; Sugrue and Hay, 1991). Today it is the protein or peptide preparations are pure, as minor thought that once a virion has entered the cell by endocy- chemical contaminants (e.g., low concentrations of deter- tosis and it is in the acidic environment of the endosomal gent) can cause membrane perturbations and spurious lumen, the virion-associated M2 ion channel permits the conductances between the chambers of the bilayer ap- passage of protons across the virion membrane into the paratus. Recordings of membranes modified by peptides virion core. The lowered intravirion pH is presumed to such as cecropins or magainins show a variety of ampli- weaken protein–protein interactions between the matrix tudes over a wide range, noisy baselines, and low ionic (M1) protein and the ribonucleoprotein (RNP) core (re- selectivities (Chernomordik et al., 1994; Christiansen et viewed in Helenius, 1992) and considerable evidence al., 1988; Duclohier et al., 1989) and such activity is often has been accumulated recently indicating that a low pH believed to be that of a lipidic pore. In contrast, ion chan- step is required to remove from the RNPs nel-forming peptides (e.g., alamethicin, nystatin, and (Bui et al., 1996; Zhirnov, 1992). Upon fusion of the virion gramicidin) have high selectivity and quantized ampli- envelope with the endosomal membrane, free RNPs are tudes over a narrow range (reviewed in Sansom, 1991). released which are competent for transport into the nu- A different approach toward determining that a viral cleus (Bukrinskaya et al., 1982; Martin and Helenius, protein can act as a lipidic pore or an ion channel is to 1991). In the presence of influenza virus un- look for membrane permeability changes when the pro- coating is incomplete, the M1 protein is not released tein is expressed in bacteria or mammalian cells, for from the RNPs, and the RNPs fail to enter the nucleus example, using as an indicator increased uptake from (Bukrinskaya et al., 1982; Martin and Helenius, 1991). As the growth medium of an intracellularly acting inhibitor discussed below a second role of the M2 ion channel in (Guinea and Carrasco, 1994, reviewed in Carrasco, the life cycle of influenza virus is for the M2 protein to 1995). However, these measurements are not direct mea- function during its transport through the exocytic pathway surements of ion channel activity as it is possible that and to modulate the pH of the Golgi cisternae (reviewed the action of one protein in a membrane may cause pore in Hay, 1992; Lamb et al., 1994). formation by another. The notion that the M2 protein has an ion channel In considering whether viral proteins have an ion chan- activity stemmed initially from investigation of the mode nel activity it is useful to envisage a specific role for the of action of the anti-influenza A virus drug amantadine. ion channel activity in the life cycle of the virus from the The drug inhibits the ‘‘early’’ step of uncoating of influenza viewpoint of virus replication and/or virus–cell interac- A virus and for some influenza A virus subtypes amanta- tions. Ion channels are well-known targets for specific dine also inhibits a step ‘‘late’’ in the replicative cycle inhibition by drug action and thus viral ion channel activity (reviewed in Hay, 1992; Lamb et al., 1994). Mutants of provides a well-defined target for new antiviral chemo- influenza virus resistant to amantadine contain amino therapeutics. In this review we discuss the known biologi- acid changes in the M2 protein transmembrane domain cal properties in virally infected cells and the electrophysi- suggesting an interaction of amantadine with the M2 pro- ological properties of the M2 and NB proteins of influenza tein (Hay et al., 1985). Later, it was observed that the A and B and Vpu and Vpr proteins of HIV-1. late effect of amantadine on viral replication involved the (HA). A large body of immunological and biochemical data indicated that addition of amantadine INFLUENZA A VIRUS M2 PROTEIN: BIOLOGICAL ASPECTS to cells late in infection caused HA to undergo its confor- mational change to the low pH-induced form in the trans- A major strength in the development of the case that Golgi network (TGN) (Ciampor et al., 1992a,b; Grambas the influenza A virus M2 protein has an ion channel activ- et al., 1992; Grambas and Hay, 1992; Sugrue et al., 1990). ity came from understanding the biology of viral replica- Addition of the Na//H/ ionophore monensin at low con- tion and its inhibition by the antiviral drug amantadine centrations (nanomolar) to amantadine-treated infected before electrophysiological experiments showed that the cells restored HA to its pH neutral form, without affecting

M2 protein has an ion channel activity which is inhibited HA transport, functionally overcoming the amantadine by amantadine. The influenza A virus M2 integral mem- block of the M2 ion channel and suggesting a similarity brane protein is abundantly expressed in virus-infected in mechanism of monensin and the M2 ion channel cells (Lamb and Choppin, 1981; Lamb et al., 1985; Zeb- (Grambas et al., 1992; Sugrue et al., 1990). Thus, the M2 edee et al., 1985) but it is only a minor, albeit essential, ion channel activity is thought to function in the TGN and component of virions (Zebedee and Lamb, 1988). The associated transport vesicles to keep the pH above the

M2 protein is a disulfide-linked homotetramer with each threshold at which the HA conformational change to the chain consisting of 97 amino acids, and it is thought low pH-induced form occurs (Ciampor et al., 1992a; that there are 24 N-terminal extracellular residues, 19 Sugrue et al., 1990). Further evidence that the M2 protein

AID VY 8451 / 6a2a$$$241 02-13-97 14:46:20 vira AP: Virology VIRAL ION CHANNELS 3 can alter the pH of intracellular compartments in the influenza B virus NB protein is a type III integral mem- absence of an influenza virus infection was obtained by brane protein (classification of von Heijne, 1988) with an examining the effect of M2 on the biogenesis of HA when 18-residue N-terminal extracellular domain, a 22-residue HA and M2 were expressed from cDNAs (Ohuchi et al., transmembrane domain, and a 60-residue cytoplasmic 1994; Sakaguchi et al., 1996; Takeuchi and Lamb, 1994). tail (Shaw et al., 1983; Williams and Lamb, 1986) but M2 and NB do not share identity. The NB protein

INFLUENZA A VIRUS M2 PROTEIN: is abundantly expressed at the surface of virus-infected ELECTROPHYSIOLOGICAL MEASUREMENTS cells (Shaw and Choppin, 1984; Williams and Lamb, 1986) and most importantly for its proposed role in virus Direct evidence that the M protein has ion channel 2 uncoating, NB is incorporated at low levels into virions activity was provided by expressing the M protein in 2 (Betakova et al., 1996; Brassard et al., 1996). As described either oocytes of X. laevis (Holsinger et al., 1994, 1995; above, the elucidation of the role of the influenza A virus Pinto et al., 1992; Sakaguchi et al., 1997; Shimbo et al., M protein stemmed from studies on understanding the 1996; Tu et al., 1996; Wang et al., 1993, 1995a) or in 2 mechanism of action of the antiviral drug amantadine. mammalian cells (Chizhmakov et al., 1996; Wang et al., However, influenza B virus replication is not inhibited by 1994a) and measuring cell surface currents. The M ion 2 amantadine at micromolar concentrations (Davies et al., channel activity is specifically blocked by the anti-influ- 1964) and there is no other known specific inhibitor of enza virus drug amantadine and is activated at the low- influenza B virus replication. At millimolar concentrations ered pH found intralumenally in endosomes and the amantadine inhibits the uncoating of influenza B virus TGN (Chizhmakov et al., 1996; Pinto et al., 1992; Shimbo and many other enveloped viruses that enter cells et al., 1996; Wang et al., 1995a). In addition, introduction through the endocytic pathway and need lowered pH to of purified M protein or the TM domain peptide into 2 trigger fusion of the with the endosomal planar lipid bilayers resulted in an amantadine-sensitive membrane. This is because amantadine can act as a ion channel activity that was activated by low pH (Duff lysosomaltropic agent (similar to NH Cl or chloroquine) and Ashley, 1992; Schroeder et al., 1994; Tosteson et 4 and raise the lumenal pH of intracellular compartments al., 1994). (reviewed in Mellman et al., 1986). Thus, all the accumulated data indicate that the influ- enza virus M2 is an ion channel protein and that it is a minimalistic or primative ion channel by comparison to INFLUENZA B VIRUS NB PROTEIN: the multi-membrane-spanning domain structure of many ELECTROPHYSIOLOGICAL MEASUREMENTS eukaryotic ion channels. Expression of NB in oocytes of X. laevis acts as an activator of an endogenous chloride conductance INFLUENZA B VIRUS NB PROTEIN: (Shimbo et al., 1995), and a similar current can be evoked BIOLOGICAL ASPECTS from oocytes that express a variety of cellular and viral The influenza B virus replicative cycle is very similar small integral membrane proteins (Attali et al., 1993; to that of influenza A virus and in vitro experiments sug- Shimbo et al., 1995; Tzounopoulos et al., 1995). Thus, the gest that a low pH step is required to release the matrix oocyte system which has proven extremely useful for protein from the RNPs (Zhirnov, 1990, 1992). Thus, it expression and measurement of many heterologous ion seems reasonable to believe that influenza B virions con- channels does not prove useful for examination of an ion tain an ion channel to permit ions to enter the virion channel activity intrinsic to NB. and weaken protein–protein interactions. Neither is the Ion channel activity in artificial lipid bilayers has been influenza B virus HA cleaved in the TGN nor is the pH reported after addition of purified NB protein synthesized required for HA to induce fusion (pH 5.4) (Brassard and in Escherichia coli (Sunstrom et al., 1996). Lipid bilayers Lamb, unpublished observation) above that of the pH of have been used as a model system to study the proper- the Golgi lumen. Therefore, there is not a requirement a ties of many ion channels, both ligand- and voltage- priori for an ion channel to equilibrate the pH of the gated. However, as mentioned above, the starting point lumen of the TGN with the to keep HA in its for the use of this model system is usually the known pH neutral form. Unlike RNA segment 7 of influenza A properties of the channel, determined from recordings virus which encodes the matrix (M1) and M2 ion channel made in cells. There are two reasons for this: first, pro- proteins (Lamb et al., 1981), influenza B virus RNA seg- teins and peptides containing hydrophobic domains ment 7 does not encode an integral membrane protein which are believed not to have ion channel activity in equivalent of the M2 protein (Briedis et al., 1982). How- cells are able to yield channel recordings in lipid bilayers ever, influenza B virus RNA segment 6 encodes a small (Lear et al., 1988; Tosteson et al., 1988, 1989); second, if integral membrane protein, NB (100 amino acids) (Shaw a eukaryotic integral membrane proteins is expressed in et al., 1982, 1983), and thus it is tempting to speculate E. coli it frequently leads to folding and oligomerization that NB has an ion channel activity. As with M2 protein, problems (reviewed in Doms et al., 1993) and it cannot

AID VY 8451 / 6a2a$$$241 02-13-97 14:46:20 vira AP: Virology 4 LAMB AND PINTO be determined if the TM domain is inserted into bilayers with an N-terminal 24-residue hydrophobic membrane- in its natural manner. Antibody to the NB C-terminus spanning domain and a C-terminal cytoplasmic tail inhibited channel activity suggesting that the activity was (Maldarelli et al., 1993) which forms homooligomers of due to NB and not a contaminant (Sunstrom et al., 1996). an undetermined number of subunits (Maldarelli et al., Nonetheless, the data must be treated cautiously be- 1993). The Vpu cytoplasmic tail is specifically phosphory- cause there is no a priori information about the NB chan- lated on serine residues 52 and 56 (Friborg et al., 1995; nel and it is possible that the ion channel activity re- Schubert et al., 1994; Schubert and Strebel, 1994). As corded in the bilayer experiments was not representative with M2 and NB of influenza A and B viruses, Vpu is a of the function of the NB protein in either the virion or type III integral membrane protein. However, unlike M2 infected cells. One of the strengths of this model system and NB proteins which are abundantly expressed at the is the ability to detect currents that result from a single cell surface (Lamb et al., 1985; Williams and Lamb, 1986), macromolecular complex, but this strength can also Vpu is largely expressed intracellularly (Klimkait et al., cause difficulty in the interpretation of results, as the 1990; Schubert et al., 1996a) and in an HIV-1 infection recorded activity cannot be attributed definitively to the Vpu is not detected at the cell surface, a finding which majority protein without substantiating evidence such as may account for its seeming absence from virions that provided by application of a specific inhibitor. The (Strebel et al., 1989). Nonetheless, when Vpu is overex- bilayer experiments did demonstrate that 2–3 mM aman- pressed, using expression vectors, some Vpu is ex- tadine inhibited the measured permeability (Sunstrom et pressed at the cell surface (Maldarelli and Strebel, un- al., 1996). However, such high concentrations of amanta- published observations cited in Schubert et al., 1996). dine have also been shown to inhibit the nicotonic acetyl- Vpu is designated an HIV-1 accessory protein because choline receptor, the NMDA (N-methyl-D-aspartate) re- its can be deleted from HIV-1 without completely ceptor channel, and ATP-regulated K-currents in HIT-T15 abrogating virus replication in vitro (Cohen et al., 1988; cells (Albuquerque et al., 1978; Ashcroft et al., 1991; Bre- Klimkait et al., 1990; Strebel et al., 1988, 1989). However, sink et al., 1995; Warnick et al., 1982). expression of Vpu has been shown to enhance virus In separate experiments Ewart and co-workers (1996) particle release from a wide variety of cell types (Schu- investigated whether NB permitted H/ to permeate, using bert et al., 1995; Strebel et al., 1988; Terwilliger et al., a NADH-dependent quinacrine fluorescence-quenching 1989). It was found both in A3.01 cells and HeLa cells assay and membrane vesicles prepared from E. coli ex- transiently expressing CD4 that there was a two- to four- pressing NB protein. The vesicles showed a reduced fold increase in particle secretion in cultures infected quinacrine fluorescence-quenching activity compared to with wt HIV-1 compared to cultures infected with a vpu- control E. coli, suggesting that the vesicles were unable deficient HIV-1 (Schubert et al., 1995, 1996a; Strebel et al., / / to maintain a high H in/H out ratio due to a proton flux 1989). With vpu-deficient HIV-1, a significant proportion of through an NB channel. However, such measurements the virions remains associated with the cell surface and are indirect and could reflect flux through a preexisting there is the formation of intracytoplasmic particles, often or NB-induced E. coli-specific channel. It is unfortunate of aberrant morphology, contained within membranous that the M2 protein was not tested in parallel, as it is know vacuoles (Klimkait et al., 1990). Interestingly, vortexing of to have a proton flux and can be inhibited by micromolar vpu-deficient HIV-1-infected cells caused a large propor- concentrations of amantadine. Nonetheless, there is still tion of the cell-associated particles to be released, sug- a risk of correlating directly an effect with a causative gesting that they are held to cells by a loose association mechanism because we note that when influenza virus (Klimkait et al., 1990). It was also observed that vpu-

M2 protein was expressed in yeast, the amantadine-sen- deficient HIV-1 caused a considerably greater cytopathic sitive M2-specific proton flux caused acidification of the effect than wt HIV-1, perhaps due to the intracellular extracellular medium possibly due to a disruption of the accumulation of possibly toxic HIV-1 proteins (Klimkait plasma membrane H/-ATPase-generated electrochemi- et al., 1990; Schubert et al., 1995; Yao et al., 1992). cal gradient (Kurtz et al., 1995). A second function of Vpu is to induce the specific Our present understanding of the replication of influ- degradation of CD4, one of the two HIV-1 coreceptor enza B virus leads us to expect that the NB protein func- molecules, during transport of CD4 through the exocytic tions as an ion channel. However, a great deal of work pathway (Lenberg and Landau, 1993; Vincent et al., 1993; needs to be done to extrapolate the observed bilayer Willey et al., 1992a,b). A significant proportion of the HIV- data for NB ion channel activity to an activity of the NB 1 envelope precursor gp160 forms a stable protein in virus-infected cells or virions. complex with CD4 and this complex is trapped in the (ER) (Bour et al., 1991; Buonocore HIV-1 Vpu: BIOLOGICAL ASPECTS and Rose, 1990; Crise et al., 1990; Jabbar and Nayak, 1990). Release of gp160 from an intracellular trap is pre- The HIV-1 is an 81-residue integral mem- dicted to increase the infectivity of HIV-1 virions. In the brane protein (Cohen et al., 1988; Strebel et al., 1988), presence of Vpu, the stability of CD4 is reduced 30- to

AID VY 8451 / 6a2a$$$241 02-13-97 14:46:20 vira AP: Virology VIRAL ION CHANNELS 5

40-fold (Willey et al., 1992b). It is thought likely that degra- is not thought that is the functional equivalent of dation of CD4 in the ER utilizes the same pathway as Vpu, as the two proteins are structurally dissimilar. In used by cells to degrade proteins that fail to fold or an analysis of chimeric bearing the gag- oligomerize correctly. regions of retroviruses that naturally lack vpu, it was The two activities of Vpu, induction of CD4 degradation found that expression of HIV-1 Vpu augmented the re- in the ER and enhancement of virus release, involve the lease of these viruses, suggesting that the function of two distinct structural domains of the protein, the cyto- Vpu is not restricted to HIV-1 (Gottlinger et al., 1993). plasmic tail and the TM domain, respectively, and the two Recent experiments have shown that of HIV-2 has activities occur in different intracellular compartments the ability to regulate positively HIV-2 particle release (Schubert et al., 1996a; Schubert and Strebel, 1994). For (Bour et al., 1996; Bour and Strebel, 1996; Ritter et al., Vpu-mediated degradation of CD4 the Vpu cytoplasmic 1996) and that if HIV-2 Env is expressed in trans with a tail must be specifically phosphorylated on serine resi- HIV-1 vpu-deficient mutant, HIV-2 Env restored the en- dues 52 and 56 (Schubert et al., 1996a; Schubert and hancement of virus released (Bour and Strebel, 1996). Strebel, 1994), and the Vpu cytoplasmic tail interacts Thus, it appears that Env of HIV-2 can functionally re- physically with the cytoplasmic tail of CD4 (Bour et al., place Vpu to enhance the rate of particle release. HIV-2 1995). This interaction releases gp160 from the gp160/ may not need the Vpu function of causing release of Env CD4 complex, allowing transport of gp160 to the cell from CD4 in the ER because the affinity of HIV-2 gp140 surface and initiating CD4 degradation in the ER. The for CD4 is 10- to 200-fold lower than that of HIV-1 gp160 conclusion that the Vpu TM domain was involved in viral (Bahraoni et al., 1992; Ivey-Hoyle et al., 1991; Mulligan release function was determined from experiments using et al., 1992). a mutant, VpuRD , which contained a scrambled TM do- It has been found that two regions of HIV-1 cause main sequence that was predicted to maintain an a- pore formation in planar membranes or E. coli. The do- helical structure. VpuRD was found to be incorporated mains of gp41 are the TM domain and residues 828 to normally into membranes, was able to form homooligo- 848 located in the C-terminus of the cytoplasmic tail, a mers, and was expressed stably and localized in a peri- region predicted to form an amphipathic helix (Arroyo et nuclear distribution like wt Vpu. Whereas Vpu was ca- RD al., 1995; Chernomordik et al., 1994). It has been sug- pable of binding to CD4 and inducing CD4 degradation, gested that the C-terminal region of HIV-1 gp41 may inter- incorporation of the Vpu mutation into HIV-1 yielded a RD act with the infected T to create a pore phenotype like that of vpu-deficient HIV-1 in which virions which in turn increases membrane permeability, leading were poorly released from infected cells (Schubert et al., to sodium and calcium flux, osmotic swelling, and T-cell 1996a). Vpu-mediated enhancement of virus release was necrosis (Chernomordik et al., 1994). The TM domain partially dependent on of the Vpu cyto- pore-forming capability could also be involved in HIV- plasmic tail (a Vpu phosphorylation deficient HIV-1 induced cytopathology (Arroyo et al., 1995) or it could be yielded a 50–100% increase in particle release over vpu- involved in forming a fusion pore during gp41-mediated deficient HIV-1) (Schubert et al., 1996a; Schubert and virus–cell fusion. It is currently unclear if the HIV-1 gp41 Strebel, 1994), suggesting that phosphorylation of the Vpu cytoplasmic tail can modify the function of the TM pore-forming ability is related to the ability of HIV-2 gp140 domain. It is interesting to note that 1H NMR and CD to functionally replace Vpu. spectroscopy studies indicate that the Vpu cytoplasmic Based on the overall similarity of structure of Vpu to domain consists of two amphipathic a-helices and these the influenza A virus M2 protein and the need to explain two a-helices are joined by a flexible highly acidic inter- the mechanism of action of the two functions of Vpu, connecting hinge region containing both phosphoaccep- Willey and co-workers (1992a) rationalized that Vpu might tor sites (Federau et al., 1996; Wray et al., 1995). Thus, have an ion channel function. Once the data using VpuRD it is possible that the phosphorylated region of Vpu might were obtained it was proposed that the Vpu TM domain act as a hinge domain and undergo conformational forms an ion conductive membrane pore (Schubert et changes that regulate biological activity. Phosphorylation al., 1996a). It was envisaged that Vpu could be directly is thus likely to regulate both activities of the Vpu pro- involved in virus release at the cell surface, although little teins, including TM domain-mediated enhancement of Vpu is expressed at the cell surface. An alternative idea virus release. also proposed was that the postulated Vpu ion channel Although HIV-1 encodes Vpu, none of the other primate activity would function to prevent intracellular budding of or nonprimate encode a Vpu-like protein, ex- virus (Schubert et al., 1996a). Given that HIV-2 Env can cept for the chimpanzee simian immunodeficiency virus functionally replace Vpu to enhance the rate of HIV-1 (Huet et al., 1990). The main difference in genome organi- particle release, the inference is that HIV-2 Env would zation between HIV-1 and HIV-2 is the presence of vpu also have an ion channel activity, presumably involving in HIV-1, which is absent from HIV-2, and the presence the TM and/or lengthy cytoplasmic tail of gp41 (Bour et of vpx in HIV-2, which is absent from HIV-1. However, it al., 1996; Ritter et al., 1996).

AID VY 8451 / 6a2a$$$241 02-13-97 14:46:20 vira AP: Virology 6 LAMB AND PINTO

HIV-1 Vpu: ELECTROPHYSIOLOGICAL tion of the cytoplasmic tail would cause a gain of function MEASUREMENTS of an ion channel activity over phosphorylated wt Vpu. An inherent difficulty with Vpu is that it lacks an ectodomain To test if Vpu had an ion channel activity it was ex- recognizable by an antibody, thus making it difficult to pressed in E. coli as a glutathione S- (GST) show if Vpu is expressed at the surface of oocytes; sur- , released by thrombin cleavage, and puri- face expression is a prerequisite for whole-cell channel fied by HPLC anion-exchange chromatography (Ewart et recordings. Indeed, as discussed above, very little wt al., 1996). After applying Vpu to bilayers, variable currents Vpu is thought to be transported to the plasma membrane of over a 20-fold range of amplitude were recorded and of mammalian cells. To attempt to show that the Vpu TM they showed permeability to Na/,K/,Cl0, and phosphate domain is the ion conductive pore of a channel, Schubert ions. The combination of a wide range of amplitude and and co-workers (1996b) constructed a chimeric mole- permeability is consistent with a generalized membrane cule, VpuR26 , in which the Vpu TM domain sequence was perturbation (see Lear et al., 1988). As discussed above randomized and the cytoplasmic tail was derived from for the studies with NB, the lipid bilayer system serves Vpu2/6 . The VpuR26 randomized hydrophobic domain still well as a model for study of ion channels that are already acts as a transmembrane domain (Schubert et al., 1996a) well-characterized, but in attempting to characterize yet when VpuR26 was expressed in oocytes even at high channels de novo there is the risk that the bilayer system levels, no surface current was detected. When VpuR26 may report ion channel activity of proteins that are not was incorporated into the HIV-1 genome it yielded a phe- ion channels. This is of particular concern when the fold- notype that is indistinguishable from vpu-deficient HIV- ing and oligomerization state of a protein synthesized 1. Thus, there is an interesting correlation of findings in and purified from bacteria have not been evaluated. that the VpuR26 mutation has an effect on the ion channel As a separate line of investigation it was found that activity induced by expression of Vpu and also on the an E. coli proline auxotroph could be made to grow under biology of HIV-1 replication. conditions that required the dissipation of a Na/ gradient Schubert and co-workers (1996b) extended their stud- of adjacent E. coli cells expressing the Vpu protein and ies by examining the ion channel-forming ability of the the consequent leakage of proline from these cells Vpu TM domain peptide. When studied under similar (Ewart et al., 1996). However, this result does not indicate conditions to those used by Ewart et al. (1996) for the definitively that the Vpu protein formed a Na/ channel in Vpu protein, bilayers exposed to the Vpu TM peptide also the adjacent cells, as the assay would also yield a posi- showed a range of currents, but in this case the range tive result if the Vpu protein acted as the auxiliary subunit of amplitude was less, 3:1, and increased in clearly iden- of a cellular protein, the same difficulty that has plagued tifiable quantized steps; the ion selectivity was also cat- some expression studies in oocytes (Attali et al., 1993; ionic for the peptide. Unlike the whole-cell currents of Shimbo et al., 1995; Tzounopoulos et al., 1995). oocytes, bilayers exposed to Vpu peptide did not show Schubert and co-workers (1996b) expressed the Vpu activation under similar conditions by voltage. The heter- cytoplasmic tail phosphorylation-deficient mutant Vpu2/6 ogeneity of the size of the conductances presumably in oocytes of X. laevis and using a two-electrode voltage arises from the nonordered self-assembly of the TM do- clamp apparatus observed small whole-cell inward cur- main peptides. A peptide (Vpurd) composed of a random- rents in response to hyperpolarizing membrane voltages ized sequence derived from the Vpu TM domain (and the

(600 nA at 130 mV). The time course of the current was same sequence as found in the TM domain of VpuR26) did characterized by slow activation by large negative mem- not form ion channels but exhibited erratic fluctuations in brane voltages, not usually encountered in mammalian membrane currents with variable amplitudes reflecting cells. Examination of the ion selectivity of the Vpu-in- transient disruptions of the membrane. duced conductance indicated permeability to monova- Taken together, the data from Ewart and co-workers lent cations without discriminating Na/ over K/ and the (1996) and Schubert and co-workers (1996b) are sugges- conductance was pH independent. Vpu2/6 was thought tive that Vpu has a channel activity. The correlation of not to activate a chloride channel endogenous to oocytes the lack of channel activity of both VpuR26 in oocytes and that can be activated by other small cellular and viral the peptide Vpurd in bilayers with the biological observa- membrane proteins (Attali et al., 1993; Shimbo et al., tion of the lack of enhancement of virus release with

1995; Tzounopoulos et al., 1995) based on inhibitor stud- HIV-1 expressing VpuR26 is important. Because electrical ies (Schubert et al., 1996b); however, in the absence of recordings from oocytes and planar bilayers can lead to a specific Vpu channel activity blocker, it cannot be ruled ambiguities and artifacts, to show without any doubt that out that Vpu does not activate an endogenous cation Vpu has an ion channel activity relevant to the HIV life channel. Unfortunately, wt Vpu did not accumulate in cycle it will be necessary to make electrophysiological oocytes and thus only very low or background level mem- recordings from mammalian cells, make mutants in the brane currents were recorded from oocytes expressing Vpu TM domain that alter ionic selectivities to show that wt Vpu, but it seems unlikely that the lack of phosphoryla- the Vpu TM domain conducts an ionic flux, reconstitute

AID VY 8451 / 6a2a$$$241 02-13-97 14:46:20 vira AP: Virology VIRAL ION CHANNELS 7 the ion channel activity of properly folded and oligo- distinct mechanisms by which HIV-1 preintegration com- merized Vpu purified from eukaryotic cells and identify plexes enter the nucleus: an importin-independent path- an inhibitor that blocks channel activity so that the speci- way mediated by Vpr (Vpr lacks a conventional nuclear ficity of channel activity can be readily identified, and localization signal) (Gallay et al., 1996) and an importin- also to obtain inhibitor-resistant mutants that map to vpu. dependent pathway mediated in part by the MA protein It is also going to require considerable effort to determine (Bukrinsky et al., 1993; Gallay et al., 1995; von Schwedler how the ion channel activity observed in vitro relates et al., 1994). to and/or causes the known biological action of Vpu in T lymphocytes infected with HIV-1 that contain an in- mediating HIV release. This is because the channel activ- tact vpr gene do not become chronically infected but ity requires a large negative membrane voltage to be ultimately die (He et al., 1995; Mustafa and Robinson, activated, not usually found in cells, and, when induced, 1993; Planelles et al., 1995; Rogel et al., 1995), apparently permits the flow of a large range of ions. because expression of Vpr causes cells to accumulate in the G2 phase of the cell cycle (Bartz et al., 1996; Jowett HIV-1 Vpr: BIOLOGICAL ASPECTS et al., 1995; Re et al., 1995; Rogel et al., 1995). This is a conserved function of human and primate lentiviruses The HIV-1 Vpr protein is a 96-residue basic protein (Planelles et al., 1996). Some tumor cell lines differentiate containing no uncharged regions longer than 8 residues; as a result of the block to cell proliferation (Levy et al., thus, it lacks a traditional membrane-spanning domain. 1993). Vpr arrests the cell cycle in G2 by preventing the Vpr is found as a component of the virion (Cohen et al., activation of the p34cdc2/cyclin B complex. It is thought 1990; Wang et al., 1994b). Vpr is another HIV-1 auxiliary that Vpr must act upstream of p34cdc2 activation by de- protein, which is not essential for replication in tissue phosphorylation as Vpr expression causes p34cdc2 to re- culture but has two confirmed properties. (1) Vpr plays main in the phosphorylated, inactive state (Di Marzio et a role in permitting entry of the viral core into the nucleus al., 1995; He et al., 1995; Jowett et al., 1995; Re et al., of nondividing cells and (2) Vpr prevents the establish- 1995). A low level of activation of driven by ment in vitro of chronically infected HIV-1 producer cell the viral LTR has been observed to be mediated by Vpr lines by causing infected cells to arrest in the G2/M (Levy et al., 1993) but this may be a result of cell-cycle phase of the cell cycle. As properties of Vpr have been arrest in G2 (Emerman, 1996). Biochemical and genetic extensively reviewed very recently (Emerman, 1996), only experiments to seek proteins that interact with Vpr have a brief synopsis will be provided here. yielded the transcription factor Sp 1 (Wang et al., 1995b), Retroviruses depend on cell division for their replica- a 41-kDa cytosolic protein that also coimmunoprecipi- tion because breakdown of the nuclear envelope at mito- tates with the glucocorticoid receptor (Refaeli et al., 1995) sis allows the viral preintegration complex to interact and uracil DNA– (Bouhamdan et al., 1996). with the host cell . In contrast, HIV-1 can However, it is not understood how these proteins can infect nondividing cells, a factor essential for infection be related to the process by which Vpr prevents cells of terminally differentiated macrophages. Vpr plays a from passing into mitosis. role in the process by linking the preintegration complex with the cell nuclear import pathway (Heinzinger et al., HIV-1 Vpr: ELECTROPHYSIOLOGICAL 1994). A minor phosphorylated form of the viral matrix MEASUREMENTS (MA) protein also participates in this process and in this regard the functions of MA and Vpr are redundant. In As discussed above, in mammalian cells Vpr is a solu- virus-infected cells Vpr accumulates largely in the nu- ble protein that is transported into the nucleus and no cleus (Lu et al., 1993; Mahalingam et al., 1995). A crude biochemical evidence has been obtained to indicate that fractionation of HIV-1-infected cells into nuclear, cyto- Vpr is an integral membrane protein. However, when Vpr solic, and membrane fractions indicated that 20% of Vpr was expressed as a GST fusion protein in E. coli, purified was associated with membranes but when Vpr was ex- by affinity chromatography and thrombin cleavage and pressed from a virus vector, the membrane- tested in planar bilayers, an ion channel activity was associated Vpr was only 8% of total (Lu et al., 1993). detected which showed a variety of amplitudes and low Established biochemical experiments to test for a pe- ionic selectivity (Piller et al., 1996). The Vpr ion channel ripheral or integral association with membranes versus activity was inhibited by antibody to Vpr, suggesting that a fractionation cross-contamination were not performed the activity observed was due to Vpr protein and not a and thus the significance of the Vpr found in the mem- contaminant (Piller et al., 1996). Although this suggests brane fraction cannot be evaluated. Vpr is incorporated that the observed currents were the result of the pres- into virions through an interaction dependent on a do- ence of the Vpr protein, it is possible that they resulted main (p6) located at the C-terminus of the Gag precursor from generalized membrane perturbation that is able to protein (Kondo et al., 1995; Lavallee et al., 1994; Lu et be modulated by antibodies. As discussed above, in the al., 1995; Paxton et al., 1993). There appear to be two absence of other supporting data, planar bilayer data

AID VY 8451 / 6a2a$$$241 02-13-97 14:46:20 vira AP: Virology 8 LAMB AND PINTO indicating ion channel activity and its relationship to a reviewers of the manuscript. Nonetheless, the opinions expressed are property of a protein in a virally infected cell must be our own. We are indebted to Drs. Schubert, Strebel, and Montal for a preprint of their manuscript and to Dr. Josh Zimmerberg, National Insti- treated very cautiously. tutes of Health, for insightful discussions. We apologize to authors who may feel sleighted because their work was not cited but the length CONCLUSIONS restriction of this review necessitated a curtailment of referencing. Research in the authors’ laboratories was supported by Public Health Do the NB, Vpu, and Vpr proteins function as ion chan- Service Research Grants AI-20201 (R.A.L.) and AI-31882 (L.H.P.) from nels in the life cycles of their respective viruses? At pres- the National Institute of Allergy and Infectious Diseases. R.A.L. is an Investigator of the Howard Hughes Medical Institute. ent, only subjective conclusions can be drawn. NB is probably an ion channel but in the large part that conclu- sion is due to our understanding of the biology of influ- REFERENCES enza virus uncoating and the parallel with influenza A Albuquerque, E. X., Eldefrawi, A. T., Eldefrawi, M. E., Mansour, N. A., and Tsai, M.-C. (1978). Amantadine: Neuromuscular blockade by sup- virus M2 ion channel activity. Vpu is possibly an ion chan- nel in the large part due to the concordance of the electro- pression of ionic conductance of the acetylcholine receptor. Science 199, 788–790. physiological data with the mutants and a loss of function Arroyo, J., Boceta, M., Gonzalez, M. A., Michel, M., and Carrasco, L. of Vpu when the same mutations are introduced into the (1995). Membrane permeabilization by different regions of the human virus genome. However, how a Vpu ion channel activity immunodeficiency virus type 1 transmembrane glycoprotein gp41. J. might mediate the release of virions is unknown. For Vpr, Virol. 69, 4095–4102. the available evidence concerning the biological role of Ashcroft, F. M., Kerr, A. J., Gibson, J. S., and Williams, B. A. (1991). Aman- tadine and sparteine inhibit ATP-regulated K-currents in the insulin- Vpr in the HIV-1 life cycle makes it seem improbable that secreting b-cell line, HIT-T15. Br. J. Pharmacol. 104, 579–584. Vpr has an ion channel activity in the virus life cycle or Attali, B., Guillemare, E., Lesage, F., Honore, E., Romey, G., Lazdunski, in virus-induced pathogenesis. However, the hypotheses M., and Barhanin, J. (1993). The protein IsK is a dual activator of K/ cannot be solely addressed by in vitro electrophysiologi- and Cl0 channels. Nature 365, 850–852. cal recordings and it would be unfortunate if efforts were Bahraoni, E., Benjouad, A., Guetard, D., Kolbe, H., Gluckman, J. C., and Montagnier, L. (1992). Study of the interaction of HIV-1 and HIV-2 expended on experiments that can never resolve the envelope with the CD4 receptor and role of N-. question definitively. If HIV-1 Vpu and Vpr are ion chan- AIDS Res. Hum. Retroviruses 8, 565–573. nels in virus-infected cells and two domains in gp41 Bartz, S., Rogel, M. E., and Emerman, M. (1996). Human immunodefi- cause lipidic pore formation, it needs to be addressed ciency virus type 1 cell cycle control: Vpr is cytostatic and mediates why this virus needs redundancy in ion channels and G2 accumulation by a mechanism which differs from DNA damage checkpoint control. J. Virol. 70, 2324–2331. pores. Other enveloped viruses, apart from HIV-1, that Betakova, T., Nermut, M. V., and Hay, A. J. (1996). The NB protein is an enter cells by fusion with the plasma membrane and bud integral component of the membrane of influenza B virus. J. Gen. from the plasma membrane, such as paramyxoviruses, Virol. 77, 2689–2694. do not seem to have an ion channel protein, although a Bouhamdan, M., Benichou, S., Rey, F., Navarro, J. M., Agostini, I., Spire, pore created by the viral fusion (F) protein probably forms B., Camonis, J., Slupphaus, G., Vigne, R., Benarous, R., and Sire, J. (1996). Human immunodeficiency virus type 1 Vpr protein binds to during virus–cell fusion. Differences in interactions be- the uracil DNA glycosylase DNA repair . J. Virol. 70, 697– tween viral proteins and cellular organelles may be key 704. to understanding virus-induced pathogenesis. Bour, S., Boulerice, F., and Wainberg, M. A. (1991). Inhibition of gp160 Ion channels are attractive targets for antiviral action, and CD4 maturation in U937 cells after both defective and productive as they have very high specific activity. A single blocking infections by human immunodeficiency virus type 1. J. Virol. 65, 6387– 6396. molecule can prevent the flow of millions of ions through Bour, S., Schubert, U., Peden, K., and Strebel, K. (1996). The envelope a single ion channel protein. The proposed ion channel glycoprotein of human immunodeficiency virus type 2 enhances viral molecules NB, Vpu, and Vpr are not homologous to particle release: A Vpu-like factor? J. Virol. 70, 820–829. known cellular proteins, thus offering the chance for Bour, S., Schubert, U., and Strebel, K. (1995). The human immunodefi- highly specific interventions. However, this promise will ciency virus type 1 Vpu protein specifically binds to the cytoplasmic domain of CD4: Implications for the mechanism of degradation. J. not be realized for these three proteins without demon- Virol. 69, 1510–1520. strating that each is a bona fide ion channel with a pro- Bour, S., and Strebel, K. (1996). The human immunodeficiency virus teinaceous pore and eliminating the possibility that these (HIV) type 2 envelope protein is a functional complement to HIV type proteins do not form lipidic pores or are membrane per- 1 Vpu that enhances particle release of heterologous retroviruses. turbants and providing further evidence linking each to J. Virol. 70, 8285–8300. Brassard, D. L., Leser, G. P., and Lamb, R. A. (1996). Influenza B virus a role in the . NB glycoprotein is a component of virions. Virology 220, 350–360. Bresink, I., Danysz, W., Parsons, C. G., and Mutschler, E. (1995). Differ- ACKNOWLEDGMENTS ent binding affinities of NMDA receptor channel blockers in various brain regions—Indication of NMDA receptor heterogeneity. Neuro- We appreciate the comments on this review made by Drs. Klaus pharmacology 34, 533–540. Strebel and Uli Schubert, National Institutes of Health, Dr. Maurice Briedis, D. J., Lamb, R. A., and Choppin, P. W. (1982). Sequence of RNA Montal, University of California, San Diego, three of Dr. Graeme Laver’s segment 7 of the influenza B virus genome: Partial amino acid homol- colleagues at the Australian National University, and the formal two ogy between the membrane proteins (M1) of influenza A and B viruses

AID VY 8451 / 6a2a$$$241 02-13-97 14:46:20 vira AP: Virology VIRAL ION CHANNELS 9

and conservation of a second open reading frame. Virology 116, Federau, T., Schubert, U., Floûdorf, J., Henklein, P., Schomburg, D., and 581–588. Wray, V. (1996). Solution structure of the cytoplasmic domain of the

Bui, M., Whittaker, G., and Helenius, A. (1996). Effect of M1 protein and human immunodeficiency virus type 1 encoded virus protein U (Vpu). low pH on nuclear transport of influenza virus ribonucleoproteins. J. Int. J. Pept. Protein Res. 47, 297–310. Virol. 70, 8391–8401. Friborg, J., Ladha, A., Gottlinger, H., Haseltine, W. A., and Cohen, E. A. Bukrinskaya, A. G., Vorkunova, N. K., Kornilayeva, G. V., Narmanbetova, (1995). Functional analysis of the phosphorylation sites on the human R. A., and Vorkunova, G. K. (1982). Influenza virus uncoating in in- immunodeficiency virus type 1 Vpu protein. J. Acq. Immune Def. fected cells and effects of rimantadine. J. Gen. Virol. 60, 49–59. Syndr. Hum. Retrovirol. 8, 10–22. Bukrinsky, M. I., Haggerty, S., Dempsey, M. P., Sharova, N., Adzhubei, Gallay, P., Stitt, V., Mundy, C., Oettinger, M., and Trono, D. (1996). Role A., Spitz, L., Lewis, P., Goldfarb, D., Emerman, M., and Stevenson, of the karyopherin pathway in human immunodeficiency virus type M. (1993). A nuclear targeting signal within HIV-1 matrix protein that 1 nuclear import. J. Virol. 70, 1027–1032. governs infection of non-dividing cells. Nature 365, 666–669. Gallay, P., Swingler, S., Aiken, C., and Trono, D. (1995). HIV-1 infection Buonocore, L., and Rose, J. K. (1990). Prevention of HIV-1 glycoprotein of non-dividing cells: C-terminal tyrosine phosphorylation of the viral transport by soluble CD4 retained in the endoplasmic reticulum. matrix protein is a key regulator. Cell 80, 379–388. Nature 345, 625–628. Gottlinger, H. G., Dorfman, T., Cohen, E. A., and Haseltine, W. A. (1993). Carrasco, L. (1995). Modification of membrane permeability by animal Vpu protein of human immunodeficiency virus type 1 enhances the viruses. Adv. Virus Res. 45, 61–112. release of produced by gag gene constructs of widely diver- Chernomordik, L., Chanturiya, A. N., Suss-Toby, E., Nora, E., and Zim- gent retroviruses. Proc. Natl. Acad. Sci. USA 90, 7381–7385. merberg, J. (1994). An amphipathic peptide from the C-terminal region Grambas, S., Bennett, M. S., and Hay, A. J. (1992). Influence of amanta- of the human immunodeficiency virus envelope glycoprotein causes dine resistance mutations on the pH regulatory function of the M2 pore formation in membranes. J. Virol. 68, 7115–7123. protein of influenza A viruses. Virology 191, 541–549. Chizhmakov, I. V., Geraghty, F. M., Ogden, D. C., Hayhurst, A., Antoniou, Grambas, S., and Hay, A. J. (1992). Maturation of influenza A virus hem- M., and Hay, A. J. (1996). Selective proton permeability and pH regula- agglutinin—Estimates of the pH encountered during transport and tion of the influenza virus M2 channel expressed in mouse erythroleu- its regulation by the M2 protein. Virology 190, 11–18. kaemia cells. J. Physiol. 494, 329–336. Guinea, R., and Carrasco, L. (1994). Influenza virus M2 protein modifies Christiansen, B., Fink, J., Merrifield, R. B., and Mazerall, D. (1988). Chan- membrane permeability in E. coli cells. FEBS Lett. 343, 242–246. nel forming properties of cecropins and related model compounds Hay, A. J. (1992). The action of adamantanamines against influenza incorporated into planar lipid membranes. Proc. Natl. Acad. Sci. USA A viruses: Inhibition of the M2 ion channel protein. Semin. Virol. 85, 5072–5076. 3, 21–30. Ciampor, F., Bayley, P. M., Nermut, M. V., Hirst, E. M., Sugrue, R. J., Hay, A. J., Wolstenholme, A. J., Skehel, J. J., and Smith, M. H. (1985). The and Hay, A. J. (1992a). Evidence that the amantadine-induced, M - 2 molecular basis of the specific anti-influenza action of amantadine. mediated conversion of influenza A virus hemagglutinin to the low EMBO J. 4, 3021–3024. pH conformation occurs in an acidic trans Golgi compartment. Virol- He, J., Choe, S., Walker, R., Di Marzio, P., Morgan, D. O., and Landau, ogy 188, 14–24. N. R. (1995). Human immunodeficiency virus type 1 protein R (Vpr) Ciampor, F., Thompson, C. A., Grambas, S., and Hay, A. J. (1992b). Regu- blocks cells in the G2 phase of the cell cycle by inhibiting p34cdc2 lation of pH by the M protein of influenza A viruses. Virus Res. 22, 2 activity. J. Virol. 69, 6705–6711. 247–258. Heinzinger, N., Bukrinsky, M., Haggerty, S., Ragland, A., Kewal Ramani, Cohen, E. A., Terwilliger, E. F., Jalinoos, Y., Proulx, J., Sodroski, J. G., V., Lee, M., Gendelman, H., Ratner, L., Stevenson, M., and Emerman, and Haseltine, W. A. (1990). Identification of HIV-1 vpr product and M. (1994). The HIV-1 Vpr protein influences nuclear targeting of viral function. J. Acq. Immune Def. Syndr. 3, 11–18. nucleic acids in non-dividing cells. Proc. Natl. Acad. Sci. USA 91, Cohen, E. A., Terwilliger, E. F., Sodroski, J. G., and Haseltine, W. A. 7311–7315. (1988). Identification of a protein encoded by the vpu gene of HIV- 1. Nature 334, 532–534. Helenius, A. (1992). Unpacking the incoming influenza virus. Cell 69, Crise, B., Buonocore, L., and Rose, J. K. (1990). CD4 is retained in the 577–578. endoplasmic reticulum by the human immunodeficiency virus type Holsinger, L. J., and Lamb, R. A. (1991). Influenza virus M2 integral mem- 1 glycoprotein precursor. J. Virol. 64, 5585–5593. brane protein is a homotetramer stabilized by formation of disulfide Davies, W. L., Grunert, R. R., Haff, R. F., McGahen, J. W., Neumayer, bonds. Virology 183, 32–43. E. M., Paulshock, M., Watts, J. C., Wood, T. R., Herman, E. C., and Holsinger, L. J., Nichani, D., Pinto, L. H., and Lamb, R. A. (1994). Influ- Hoffman, C. E. (1964). Antiviral activity of 1-adamantanamine (aman- enza A virus M2 ion channel protein: A structure–function analysis. tadine). Science 144, 862–863. J. Virol. 68, 1551–1563. Di Marzio, P., Choe, S., Ebright, M., Knoblauch, R., and Landau, N. R. Holsinger, L. J., Shaughnessy, M. A., Micko, A., Pinto, L. H., and Lamb, (1995). Mutational analysis of cell cycle arrest, nuclear localization, R. A. (1995). Analysis of the posttranslational modifications of the and virion packaging of human immunodeficiency virus type 1 Vpr. influenza virus M2 protein. J. Virol. 69, 1219–1225. J. Virol. 69, 7909–7916. Huet, T., Cheynier, R., Meyerhans, A., Roelants, G., and Wain-Hobson, Doms, R. W., Lamb, R. A., Rose, J. K., and Helenius, A. (1993). Folding S. (1990). Genetic organization of a chimpanzee related to and assembly of viral membrane proteins. Virology 193, 545–562. HIV-1. Nature 345, 356–359. Duclohier, H., Molle, G., and Spach, G. (1989). Antimicrobial peptide Ivey-Hoyle, M., Culp, J. S., Chaikin, M. A., Hellmig, B. D., Matthews, T. J., magainin I from Xenopus skin forms anion permeable channels in Sweet, R. W., and Rosenberg, M. (1991). Envelope glycoproteins from planar lipid bilayers. Biophys. J. 56, 1017–1021. biologically diverse isolates of immunodeficiency viruses have Duff, K. C., and Ashley, R. H. (1992). The transmembrane domain of widely different affinities for CD4. Proc. Natl. Acad. Sci. USA 88, 512–

influenza A M2 protein forms amantadine-sensitive proton channels 516. in planar lipid bilayers. Virology 190, 485–489. Jabbar, M. A., and Nayak, D. P. (1990). Intracellular interaction of human Emerman, M. (1996). HIV-1, Vpr and the cell cycle. Curr. Biol. 6, 1096– immunodeficiency virus type 1 (ARV-2) envelope glycoprotein gp160 1103. with CD4 blocks the movement and maturation of CD4 to the plasma Ewart, G. D., Sutherland, T., Gage, P. W., and Cox, G. B. (1996). The membrane. J. Virol. 64, 6291–6304. Vpu protein of human immunodeficiency virus type 1 forms cation- Jowett, J. B. M., Planelles, V., Poon, B., Shah, N. P., Chen, M.-L., and selective ion channels. J. Virol. 70, 7108–7115. Chen, I. S. Y. (1995). The human immunodeficiency virus type 1 vpr

AID VY 8451 / 6a2a$$$241 02-13-97 14:46:20 vira AP: Virology 10 LAMB AND PINTO

gene arrests infected T cells in the G2 / M phase of the cell cycle. Ohuchi, M., Cramer, A., Vey, M., Ohuchi, R., Garten, W., and Klenk, H.- J. Virol. 69, 6304–6313. D. (1994). Rescue of vector-expressed fowl plague virus hemaggluti- Klimkait, T., Strebel, K., Hoggan, M. D., Martin, M. A., and Orenstein, nin in biologically active form by acidotropic agents and coexpressed

J. M. (1990). The human immunodeficiency virus type 1-specific pro- M2 protein. J. Virol. 68, 920–926. tein vpu is required for efficient virus maturation and release. J. Virol. Paxton, W., Connor, T. I., and Landau, N. R. (1993). Incorporation of Vpr 64, 621–629. into human immunodeficiency virus type 1 virions: Requirement for Kondo, E., Mammano, F., Cohen, E. A., and Gottlinger, H. G. (1995). The the p6 region of gag and mutational analysis. J. Virol. 67, 7229–7237. P6gag domain of human immunodeficiency virus type 1 is sufficient Piller, S. C., Ewart, G. D., Premkumar, A., Cox, G. B., and Gage, P. W. for the incorporation of Vpr into heterologous viral particles. J. Virol. (1996). Vpr protein of human immunodeficiency virus type 1 forms 69, 2759–2764. cation-selective channels in planar lipid bilayers. Proc. Natl. Acad. Kurtz, S., Luo, G., Hahnenberger, K. M., Brooks, C., Gecha, O., Ingalls, Sci. USA 93, 111–115.

K., Numata, K.-I., and Krystal, M. (1995). Growth impairment resulting Pinto, L. H., Holsinger, L. J., and Lamb, R. A. (1992). Influenza virus M2

from expression of influenza virus M2 protein in Saccharomyces cere- protein has ion channel activity. Cell 69, 517–528. visiae: Identification of a novel inhibitor of influenza virus. Antimicrob. Planelles, V., Bachelerie, F., Jowett, J. B. M., Haislip, A., Xie, Y. P. B., Agents Chem. 39, 2204–2209. Masuda, T., and Chen, I. S. Y. (1995). Fate of the human immunodefi- Lamb, R. A., and Choppin, P. W. (1981). Identification of a second pro- ciency virus type 1 provirus in infected cells: A role for vpr. J. Virol.

tein (M2) encoded by RNA segment 7 of influenza virus. Virology 112, 69, 5883–5889. 729–737. Planelles, V., Jowett, J. B., Li, Q., Xie, Y., Hahn, B., and Chen, I. S. Y. Lamb, R. A., Holsinger, L. J., and Pinto, L. H. (1994). The influenza A (1996). Vpr-induced cell cycle arrest is conserved among primate

virus M2 ion channel protein and its role in the influenza virus life lentiviruses. J. Virol. 70, 2516–2524. cycle. In ‘‘Receptor-Mediated Virus Entry into Cells’’ (E. Wimmer, Ed.), Re, F., Braaten, D., Franke, E. K., and Luban, J. (1995). Human immuno- pp. 303–321. Cold Spring Harbor Laboratory Press, Cold Spring Har- deficiency virus type 1 arrests the cell cycle in G2 by inhibiting the bor, NY. activation of p34cdc2-cyclin B. J. Virol. 69, 6859–6864. Lamb, R. A., Zebedee, S. L., and Richardson, C. D. (1985). Influenza Refaeli, Y., Levy, D. N., and Weiner, D. B. (1995). The glucocorticoid

virus M2 protein is an integral membrane protein expressed on the receptor type II complex is a target of the HIV-1 vpr gene product. infected-cell surface. Cell 40, 627–633. Proc. Natl. Acad. Sci. USA 92, 3621–3625. Lavallee, C., Yao, X. J., Ladha, A., Gottlinger, H., Haseltine, W. A., and Ritter, G. D., Jr., Yamshchikov, G., Cohen, S. J., and Mulligan, M. J. (1996). Cohen, E. A. (1994). Requirement of the Pr55gag precursor for incor- Human immunodeficiency virus type 2 glycoprotein enhancement of poration of the Vpr product into human immunodeficiency virus type particle budding: Role of the cytoplasmic domain. J. Virol. 70, 2669– 1 viral particles. J. Virol. 68, 1926–1934. 2673. Lear, J. D., Wasserman, Z. R., and DeGrado, W. F. (1988). Synthetic am- Rogel, M., Wu, L., and Emerman, M. (1995). The human immunodefi- phiphilic peptide models for protein ion channels. Science 240, ciency virus type 1 vpr gene prevents cell proliferation during chronic 1179–1181. infection. J. Virol. 69, 882–888. Lenberg, M. E., and Landau, N. R. (1993). Vpu-induced degradation of Sakaguchi, T., Leser, G. P., and Lamb, R. A. (1996). The ion channel

CD4: Requirement of specific amino acid residues in the cytoplasmic activity of the influenza virus M2 protein affects transport through the domain of CD4. J. Virol. 67, 7238–7245. Golgi apparatus. J. Cell Biol. 133, 733–747. Levy, D. N., Fernandes, L. S., Williams, W. V., and Weiner, D. B. (1993). Sakaguchi, T., Tu, Q., Pinto, L. H., and Lamb, R. A. (1997). The active

Induction of cell differentiation by human immunodeficiency virus 1 oligomeric state of the minimalistic influenza virus M2 ion channel vpr. Cell 72, 541–550. is a tetramer. Submitted. Lu, Y. L., Bennett, R. P., Wills, J. W., Gorelick, R., and Ratner, L. (1995). Sansom, M. S. P. (1991). The biophysics of peptide models of ion chan- A leucine triplet repeat sequence (LXX)4 in p6gag is important for Vpr nels. Prog. Biophys. Mol. Biol. 55, 139–235. incorporation into human immunodeficiency virus type 1 particles. J. Schroeder, C., Ford, C. M., Wharton, S. A., and Hay, A. J. (1994). Func-

Virol. 69, 6873–6879. tional reconstitution in lipid vesicles of influenza virus M2 protein Lu, Y. L., Spearman, P., and Ratner, L. (1993). Human immunodeficiency expressed by baculovirus: Evidence for proton transfer activity. J. virus type 1 viral protein R localization in infected cells and virions. Gen. Virol. 75, 3477–3484. J. Virol. 67, 6542–6550. Schubert, U., Bour, S., Ferrer-Montiel, A. V., Montal, M., Maldarelli, F., Mahalingam, S., Khan, S. A., Murali, R., Jabbar, M. A., Monken, C. E., and Strebel, K. (1996a). The two biological activities of human immu- Collman, R. G., and Srinivasan, A. (1995). Mutagenesis of the putative nodeficiency virus type 1 Vpu protein involve two separable structural a-helical domain of the Vpr protein of human immunodeficiency virus domains. J. Virol. 70, 809–819. type 1: Effect on stability and virion incorporation. Proc. Natl. Acad. Schubert, U., Clouse, K. A., and Strebel, K. (1995). Augmentation of virus Sci. USA 92, 3794–3798. secretion by the human immunodeficiency virus type 1 Vpu protein Maldarelli, F., Chen, M.-Y., Willey, R. L., and Strebel, K. (1993). Human is cell type independent and occurs in cultured human primary mac- immunodeficiency virus type 1 Vpu protein is an oligomeric type I rophages and lymphocytes. J. Virol. 69, 7699–7711. integral membrane protein. J. Virol. 67, 5056–5061. Schubert, U., Ferrer-Montiel, A. V., Oblatt-Montal, M., Henklein, P., Martin, K., and Helenius, A. (1991). Nuclear transport of influenza virus Strebel, K., and Montal, M. (1996b). Identification of an ion channel ribonucleoproteins: The (M1) promotes export activity of the Vpu transmembrane domain and its involvement in the and inhibits import. Cell 67, 117–130. regulation of virus release from HIV-1-infected cells. FEBS Lett. 398, Mellman, I., Fuchs, R., and Helenius, A. (1986). Acidification of the 12–18. endocytic and exocytic pathways. Annu. Rev. Biochem. 55, 663–700. Schubert, U., Henklein, P., Boldyreff, B., Wingender, E., Strebel, K., and Mulligan, M. J., Ritter, G. D., Jr., Chaikin, M. A., Yamshchikov, G. V., Ku- Porstmann, T. (1994). The human immunodeficiency virus type 1 mar, P., Hahn, B. H., Sweet, R. W., and Compans, R. W. (1992). Human encoded Vpu protein is phosphorylated by casein kinase-2 (CK-2) at immunodeficiency virus type 2 envelope glycoprotein: Differential positions Ser52 and Ser56 within a predicted alpha-helix-turn-alpha- CD4 interactions of soluble gp120 versus the assembled envelope helix-motif. J. Mol. Biol. 236, 16–25. complex. Virology 187, 233–241. Schubert, U., and Strebel, K. (1994). Differential activities of the human Mustafa, F., and Robinson, H. L. (1993). Context-dependent role of hu- immunodeficiency virus type 1-encoded Vpu protein are regulated man immunodeficiency virus type 1 auxiliary in the establish- by phosphorylation and occur in different cellular compartments. J. ment of chronic virus producers. J. Virol. 67, 6909–6915. Virol. 68, 2260–2271.

AID VY 8451 / 6a2a$$$241 02-13-97 14:46:20 vira AP: Virology VIRAL ION CHANNELS 11

Shaw, M. W., and Choppin, P. W. (1984). Studies on the synthesis of come to terms with membranes. Biochim. Biophys. Acta 189, 239– the influenza B virus NB glycoprotein. Virology 139, 178–184. 242. Shaw, M. W., Choppin, P. W., and Lamb, R. A. (1983). A previously unrec- von Schwedler, U., Kornbluth, R. S., and Trono, D. (1994). The nuclear ognized influenza B virus glycoprotein from a bicistronic mRNA that localization signal of the matrix protein of human immunodeficiency also encodes the viral . Proc. Natl. Acad. Sci. USA 80, virus type 1 allows the establishment of infection in macrophages 4879–4883. and quiescent T lymphocytes. Proc. Natl. Acad. Sci. USA 91, 6992– Shaw, M. W., Lamb, R. A., Erickson, B. W., Briedis, D. J., and Choppin, 6996. P. W. (1982). Complete nucleotide sequence of the neuraminidase Wang, C., Lamb, R. A., and Pinto, L. H. (1994a). Direct measurement of gene of influenza B virus. Proc. Natl. Acad. Sci. USA 79, 6817–6821. the influenza A virus M2 protein ion channel activity in mammalian Shimbo, K., Brassard, D. L., Lamb, R. A., and Pinto, L. H. (1995). Viral and cells. Virology 205, 133–140. cellular small integral membrane proteins can modify ion channels Wang, C., Lamb, R. A., and Pinto, L. H. (1995a). Activation of the M2 ion endogenous to Xenopus oocytes. Biophys. J. 69, 1819–1829. channel of influenza virus: A role for the transmembrane domain Shimbo, K., Brassard, D. L., Lamb, R. A., and Pinto, L. H. (1996). Ion histidine residue. Biophys. J. 69, 1363–1371. Wang, C., Takeuchi, K., Pinto, L. H., and Lamb, R. A. (1993). Ion channel selectivity and activation of the M2 ion channel of influenza virus. Biophys. J. 70, 1335–1346. activity of influenza A virus M2 protein: Characterization of the aman- Strebel, K., Klimkait, T., Maldarelli, F., and Martin, M. A. (1989). Molecu- tadine block. J. Virol. 67, 5585–5594. Wang, J. J., Lu, Y.-l., and Ratner, L. (1994b). Particle assembly and Vpr lar and biochemical analyses of human immunodeficiency virus type expression in human immunodeficiency virus type 1-infected cells 1 Vpu protein. J. Virol. 63, 3784–3791. demonstrated by immunoelectron microscopy. J. Gen. Virol. 75, Strebel, K., Klimkait, T., and Martin, M. A. (1988). A novel gene of HIV- 2607–2614. 1, VPU, and its 16-kilodalton product. Science 241, 1221–1223. Wang, L., Mukherjee, S., Jia, F., Narayan, O., and Zhao, L. J. (1995b). Sugrue, R. J., Bahadur, G., Zambon, M. C., Hall-Smith, M., Douglas, A. R., Interaction of virion protein Vpr of human immunodeficiency virus and Hay, A. J. (1990). Specific structural alteration of the influenza type 1 with cellular transcription factor Sp1 and trans-activation of haemagglutinin by amantadine. EMBO J. 9, 3469–3476. viral long terminal repeat. J. Biol. Chem. 270, 25564–25569. Sugrue, R. J., and Hay, A. J. (1991). Structural characteristics of the M2 Warnick, J. E., Maleque, M. A., Bakry, N., Eldefrawi, A. T., and Albuquer- protein of the influenza A viruses: Evidence that it forms a tetrameric que, E. X. (1982). Structure–activity relationships of amantadine. I. channel. Virology 180, 617–624. Interaction of the N-alkyl analogues with the ionic channel of nicotinic Sunstrom, N. A., Premkumar, L. S., Premkumar, A., Ewart, G., Cox, G. B., acetylcholine receptor and electrically excitable membrane. Mol. and Gage, P. W. (1996). Ion channels formed by NB, an influenza B Pharmacol. 22, 82–93. virus protein. J. Membr. Biol. 150, 127–132. Willey, R. L., Maldarelli, F., Martin, M. A., and Strebel, K. (1992a). Human Takeuchi, K., and Lamb, R. A. (1994). Influenza virus M2 protein ion immunodeficiency virus type 1 Vpu protein induces rapid degradation channel activity stabilizes the native form of fowl plague virus hemag- of CD4. J. Virol. 66, 7193–7200. glutinin during intracellular transport. J. Virol. 68, 911–919. Willey, R. L., Maldarelli, F., Martin, M. A., and Strebel, K. (1992b). Human Terwilliger, E. F., Cohen, E. A., Lu, Y. C., Sodroski, J. G., and Haseltine, immunodeficiency virus type 1 Vpu protein regulates the formation W. A. (1989). Functional role of human immunodeficiency virus type of intracellular gp160–CD4 complexes. J. Virol. 66, 226–234. 1 vpu. Proc. Natl. Acad. Sci. USA 86, 5163–5167. Williams, M. A., and Lamb, R. A. (1986). Determination of the orientation Tosteson, M. T., Auld, D. S., and Tosteson, D. C. (1989). Voltage-gated of an integral membrane protein and sites of by oligo- channels formed in lipid bilayers by a positively charged segment nucleotide-directed mutagenesis: Influenza B virus NB glycoprotein of the Na-channel polypeptide. Proc. Natl. Acad. Sci. USA 86, 707– lacks a cleavable signal sequence and has an extracellular NH2- 710. terminal region. Mol. Cell. Biol. 6, 4317–4328. Tosteson, M. T., Caulfield, M. P., Levy, J. J., Rosenblatt, M., and Toste- Wray, V., Federau, T., Henklein, P., Klabunde, S., Kunert, O., Schomburg, son, D. C. (1988). The synthetic precursor specific region of pre-pro- D., and Schubert, U. (1995). Solution structure of the hydrophilic 1 parathyroid hormone forms ion channels in lipid bilayers. Biosci. region of HIV-1 encoded virus protein (U) (Vpu) by CD and H NMR Rep. 8, 173–183. spectroscopy. Int. J. Pept. Protein Res. 45, 35–43. Tosteson, M. T., Pinto, L. H., Holsinger, L. J., and Lamb, R. A. (1994). Yao, X. J. Y., Goettlinger, H., Haseltine, W. A., and Cohen, E. A. (1992). Envelope glycoprotein and CD4 independence of Vpu-facilitated hu- Reconstitution of the influenza virus M ion channel in lipid bilayers. 2 man immunodeficiency virus type 1 export. J. Virol. 66, 5119– J. Membr. Biol. 142, 117–126. 5126. Tu, Q., Pinto, L. H., Luo, G., Shaughnessy, M. A., Mullaney, D., Kurtz, S., Zebedee, S. L., and Lamb, R. A. (1988). Influenza A virus M protein: Krystal, M., and Lamb, R. A. (1996). Characterization of inhibition of 2 Monoclonal antibody restriction of virus growth and detection of M M ion channel activity by BL-1743, an inhibitor of influenza A virus. 2 2 in virions. J. Virol. 62, 2762–2772. J. Virol. 70, 4246–4252. Zebedee, S. L., Richardson, C. D., and Lamb, R. A. (1985). Characteriza- Tzounopoulos, T., Maylie, J., and Adelman, J. P. (1995). Induction of tion of the influenza virus M2 integral membrane protein and expres- endogenous channels by high levels of heterologous membrane pro- sion at the infected-cell surface from cloned cDNA. J. Virol. 56, 502– teins in Xenopus oocytes. Biophys. J. 69, 904–908. 511. Vincent, M. J., Raja, N. U., and Jabbar, M. A. (1993). Human immunodefi- Zhirnov, O. P. (1990). Solubilization of matrix protein M1/M from virions ciency virus type 1 Vpu protein induces degradation of chimeric occurs at different pH for orthomyxo- and paramyxoviruses. Virology envelope glycoproteins bearing the cytoplasmic and anchor domains 176, 274–279.

of CD4: Role of the cytoplasmic domain in Vpu-induced degradation Zhirnov, O. P. (1992). Isolation of matrix protein M1 from influenza vi- in the endoplasmic reticulum. J. Virol. 67, 5538–5549. ruses by acid-dependent extraction with nonionic detergent. Virology von Heijne, G. (1988). Transcending the impenetrable: How proteins 186, 324–330.

AID VY 8451 / 6a2a$$$241 02-13-97 14:46:20 vira AP: Virology