The H+-ATPase in the Plasma Membrane of Saccharomyces cerevisiae Is Activated during Growth Latency in Octanoic Acid-Supplemented Medium Accompanying the Decrease in Intracellular pH and Cell Viability

Saccharomyces cerevisiae plasma membrane H⁺-ATPase activity was stimulated during octanoic acid-induced latency, reaching maximal values at the early stages of exponential growth. The time-dependent pattern of ATPase activation correlated with the decrease of cytosolic pH (pH_i). The cell population used as inoculum exhibited a significant heterogeneity of pH_i, and the fall of pH_i correlated with the loss of cell viability as determined by plate counts. When exponential growth started, only a fraction of the initial population was still viable, consistent with the role of the physiology and number of viable cells in the inoculum in the duration of latency under acid stress.The biological target sites of octanoic acid in Saccharomyces cerevisiae may be related to processes of transport across membranes, particularly the plasma membrane (21). Like other weak acids at low pH, octanoic acid, a highly toxic by-product of yeast alcoholic fermentation (23) and an antimicrobial food additive (6), leads to the reduction of cytosolic pH (pH_i) due to its dissociation in the approximately neutral cytoplasm following the entrance of the undissociated toxic form into the cell by passive diffusion (5, 20, 23). It is likely that this highly liposoluble weak acid significantly affects the spatial organization of the plasma membrane, affecting its function as a matrix for enzymes and as a selective barrier, thereby leading to the dissipation of the proton motive force across the plasma membrane and to intracellular acidification (16, 18). Significantly, the H⁺-ATPase in the plasma membrane in yeast, which creates the electrochemical proton gradient that drives the secondary transport of solutes and is implicated in the maintenance of pH_i around neutrality, has been pointed out as a critical component of yeast adaptation to weak acids (8, 19, 24). Indeed, yeast plasma membrane H⁺-ATPase is activated during exponential growth with octanoic acid (19, 24), and the duration of lag phase before yeast cells enter exponential growth in the presence of sorbic acid is significantly extended in a mutant with reduced levels of plasma membrane ATPase activity (8). The activation of the H⁺-ATPase in the plasma membrane in yeast cells exposed to other stresses that also lead to the dissipation of the H⁺ gradient and intracellular acidification (such as subcritical inhibitory concentrations of ethanol [12, 14, 15], supraoptimal temperatures below 40°C [25], presence of other organic acids at low pH [1, 5, 8], and deprivation of nitrogen source [2]) have also been observed. Several lines of evidence indicate that ATPase activation is due to posttranslational modifications of the PMA1 ATPase (2, 12, 24, 25). Considerable information has been obtained on the variation of plasma membrane ATPase activity during exponential growth and early stationary phase of yeast cells cultivated in media, at low pH, supplemented or not with octanoic acid (24). However, this is not the case during the period of latency preceding exponential growth at concentrations of octanoic acid close to the maximal concentration allowing growth. The main objective of the present work was to obtain information about the pattern of ATPase activity and the changes in pH_i and cell viability during the lag phase necessary for yeast adaptation to the physiological effects of octanoic acid before exponential growth.

Duration of yeast growth latency in octanoic acid-supplemented media.

When cells of S. cerevisiae IGC3507III grown, at 30°C, in medium that had not been supplemented with octanoic acid were used to inoculate buffered YG media (30 g of glucose liter⁻¹, 6.7 g of Yeast Nitrogen Base [Difco] liter⁻¹) (pH 4.0) supplemented with increasing concentrations of this toxic acid up to around 0.35 mM total acid (19, 23), exponential growth was initiated without significant delay (Fig. ​(Fig.1a),1a), although a dose-dependent decrease in specific growth rate was observed (Fig. ​(Fig.1b).1b). However, for higher concentrations up to the maximal that allowed growth (0.42 mM), a lag phase was observed and its duration strongly increased with the severity of octanoic acid stress (Fig. ​(Fig.1a).1a). The duration of latency was drastically reduced when exponential cells used as inoculum were grown in medium with an identical concentration of octanoic acid (Fig. ​(Fig.1a),1a), but the specific growth rate was not modified (Fig. ​(Fig.1b).1b). At a concentration of total octanoic acid of 0.39 mM, a lag phase of around 55 h was necessary for yeast cells, which had been cultivated under nonstressing conditions, to adapt to the deleterious effects of octanoic acid and to initiate inhibited exponential growth (Fig. ​(Fig.2).2). Open in a separate windowFIG. 1Effect of the addition to the growth medium of increasing concentrations of octanoic acid on the duration of lag phase (a) and the specific growth rate of S. cerevisiae IGC 3507III (b) for exponentially growing cells (used as inoculum) cultivated at 30°C at pH 4.0 in the absence (○) or presence (•) of concentrations of toxic lipophilic acid identical to those present in the growth medium. Results are representative of the many growth experiments carried out.Open in a separate windowFIG. 2Specific activity of plasma membrane H⁺-ATPase (filled symbols) and growth curve (open symbols) of S. cerevisiae IGC 3507III during cultivation in the presence (a) or absence (b) of 0.39 mM total octanoic acid (at pH 4.0, 30°C). The data are averages with standard deviations for at least three enzyme assays using cells from at least two independent growth experiments. OD, optical density.

Activation of plasma membrane ATPase during octanoic acid-induced latency.

The specific activity of plasma membrane ATPase assayed in crude membrane suspensions prepared from nonadapted cells, as previously reported (19, 25), during cultivation in buffered medium (at pH 4.0) supplemented with 0.39 mM octanoic acid, increased during the 55 h of latency (Fig. ​(Fig.2a).2a). A peak of activity was reached during the early stages of exponential growth and values of ATPase activity were consistently higher (twofold) in cells grown under octanoic acid stress (Fig. ​(Fig.2),2), as described by Viegas et al. (24). Yeast cells must adapt to the physiological effects of octanoic acid during an extended lag period, the length of which depended on the severity of acid stress (Fig. ​(Fig.1a),1a), before eventually recovering and entering exponential growth; the activation of plasma membrane H⁺-ATPase observed during this period of latency reinforces the idea that this proton pump is an important component of this adaptative response (5, 8, 19, 24). In fact, the ability of yeast cells to grow in the presence of lipophilic acids at a low pH reflects their capacity to maintain control over their internal pH by excluding protons. This adaptative phenomenon, reported for the first time in the present work, complements the observation of Holyoak et al. (8) that a strain with reduced plasma membrane H⁺-ATPase activity displayed increased lag phase in the presence of the weak-acid preservative sorbic acid. Significantly, plasma membrane H⁺-ATPase activity was also pointed out to play a critical role in yeast tolerance of ethanol (15) or supraoptimal temperatures (13, 25). The mechanism underlying plasma membrane ATPase activation during octanoic acid-induced latency remains obscure at the present time, but it is likely that this is due to a posttranslational modification of ATPase, as proposed for ATPase activation during octanoic acid-stressed exponential growth (24). It is likely that during lag phase the amount of H⁺-ATPase in the plasma membrane slightly decreases, as found by Benito et al. (2) in yeast cells deprived of nitrogen source where ATPase activation also occurred (2), as the estimated half-life of the enzyme is about 11 h (2). ATPase activation during latency can hardly be attributed to the adaptative modification of the ATPase lipid environment in cells grown under lipophilic acid stress, as suggested by Alexandre et al. (1).

Changes in yeast pH_i and viability during octanoic acid-stressed cultivation.

The change in pH_i during cultivation of nonadapted cells with 0.39 mM octanoic acid was monitored by using an adaptation of the fluorescence microscopic image processing technique developed by Imai and Ohno (9); 5- (and 6)-carboxyfluorescein (cF) was used as the internal pH-dependent fluoroprobe. Cells washed and resuspended in cold CF buffer (citrate-phosphate buffer [at pH 4.0] with 50 mM glycine [Sigma], 110 mM NaCl, 5 mM KCl, and 1 mM MgCl₂) to a cellular density of 2 × 10⁸ ml⁻¹ were loaded with cF by adding 20 μM of 5 (and 6)-carboxyfluorescein-diacetate (Sigma) and vortexing in two bursts of 1 min each, interspersed with 15 min on ice (9). After being washed twice with cold CF buffer, cF-loaded cells were immediately examined with a Zeiss Axioplan microscope equipped with adequate epifluorescence interference filters (Zeiss BP450-490 and Zeiss LP520) and connected to a video camera and to a computer with an image- analysis program (gel documentation system SW2000; UVP, San Gabriel, Calif.). Following a cell-by-cell analysis, the value of fluorescence intensity (fI) emitted by each cell, measured by direct densitometry, corresponded to the arithmetical mean value of fI measured in two or three different regions in the cytoplasm of the same cell, with the less fluorescent vacuole excluded. To estimate average pH_i, an in vivo calibration curve was prepared (Fig. ​(Fig.3)3) by using cell suspensions grown in the absence of toxics which were loaded with cF as described above and incubated, at 30°C, with 0.5 mM carbonyl cyanide m-chlorophenylhydrazone (CCCP) to dissipate the plasma membrane pH gradient (4), before adjustment of external pH (in the range 3.5 to 7.5) by the addition of HCl or NaOH at 2 M. Fluorescence images were fixed 15 s after the occurrence of the excitation radiation in order to minimize interferences due to leakage of cF as well as fluorescence quenching (3, 7). Cells were kept on ice throughout the procedure, and CF buffer lacked glucose; therefore, the active efflux of cF (3) was minimized as confirmed by measuring the fluorescence in the medium surrounding the cells, which was negligible. Under the experimental conditions used and for the purpose of the study, this technique proved to be highly useful and suitable despite the limitations that might be raised (3, 7). It allowed a clear-cut picture of the pH_i of individual cells, giving information about the distribution of pH_i values of a yeast population (Fig. ​(Fig.44 and ​and5a5a to c), instead of solely an estimation of the average value of the whole population, as is the case with techniques based on the distribution of radioactive propionic acid (20) or on the in vivo ³¹P nuclear magnetic resonance (5). Moreover, values calculated for the average pH_i of the whole yeast population during latency and exponential growth in medium with octanoic acid (Fig. ​(Fig.5d)5d) were close to, although slightly lower than, the values previously obtained based on the distribution of [¹⁴C]propionic acid (20, 22). Results revealed that the cell population used to inoculate octanoic acid-supplemented medium exhibited a significant heterogeneity (Fig. ​(Fig.4);4); around 31% showed a pH_i in the optimal range (above 6.5) (Fig. ​(Fig.4),4), with the average pH_i value of the whole population estimated to be approximately 6.0. This low pH_i value results from cell cultivation in a rich medium with high production of organic acids (11) (external pH, 3.6), followed by washing of the cells with YG medium buffered at pH 4.0 (17). The introduction of the inoculum in octanoic acid-supplemented medium led to the very rapid (5-min) increase in the percentage of the cell population with pH_i below 5.5, consistent with the rapid kinetics of cytosol acidification when yeast cells are exposed to weak acids (5). During extended incubation with octanoic acid and until the end of latency, the percentage of the population with a very low pH_i (below 5.5) continued to increase, reaching 80% of the cell population, while the percentage of cell population with a pH_i above 6.0 suffered a corresponding decrease (Fig. ​(Fig.5).5). During exponential growth, the opposite pH_i modification was observed, consistent with a recovery of pH_i to physiological levels (Fig. ​(Fig.5).5). The time-dependent pattern of internal acidification during lag phase correlated with plasma membrane ATPase activation (Fig. ​(Fig.2a2a and ​and5),5), suggesting that this activation was triggered by intracellular acidification, as proposed for acetic acid (5)- or nitrogen starvation (2)-induced activation. Immediately before yeast cells entered exponential growth, 80% of the initial viable population had lost viability, as assessed by the number of CFU (21) (Fig. ​(Fig.6),6), suggesting that octanoic acid-induced death during latency is related to internal acidification down to critical values (Fig. ​(Fig.55 and ​and6),6), in agreement with the relationship established by Imai and Ohno (10) between yeast viability and intracellular pH. Only about 20% of the initial population was able to start cell division in octanoic acid-supplemented medium, presumably those cells that in the inoculum exhibited pH_i values around neutrality (Fig. ​(Fig.55 and ​and6).6). These results suggest that despite plasma membrane H⁺- ATPase activation, this system of pH homeostasis may not be able to fully counteract the physiological effects of increasing octanoic acid concentrations and eventually fails at very severe acid stress. Open in a separate windowFIG. 3In vivo calibration curve, showing the pH dependence of the fI of cF-loaded-cells of S. cerevisiae IGC 3507III. Intracellular and extracellular pHs were equilibrated by incubation of cF-loaded cells, for 10 min at 30°C, with 0.5 mM CCCP. At each pH, values of fI correspond to the average fI of about 20 cells. The data are averages with standard deviations for three independent experiments.Open in a separate windowFIG. 4Distribution of cells with different pH_i values present in the inoculum of S. cerevisiae IGC 3507III prepared in growth medium without octanoic acid supplementation.Open in a separate windowFIG. 5Percentage of yeast cells with pH_i below 5.5 (a), between 5.5 and 6.0 (b), or above 6.0 (c); average pH_i of the whole cell population (▴) during S. cerevisiae IGC3507III cultivation in medium supplemented with 0.39 mM total octanoic acid (pH 4.0, 30°C); and the optical density (OD) of the culture at 600 nm (▪). The average pH_i values estimated for the whole cell population are the arithmetical mean values of the various average pH_i values calculated for individual cells. The percentage of cells present in the inoculum with pH_i values within the three ranges (○) and the average pH_i of the inoculum cell population (▵) are indicated.Open in a separate windowFIG. 6Concentration of viable cells (▴) and culture optical density (O.D.) at 600 nm (□) during lag and exponential phases of S. cerevisiae IGC 3507III growth in medium supplemented with 0.39 mM octanoic acid, at pH 4.0 and 30°C.

Adaptative response to octanoic acid.

The adaptation of yeast cells to octanoic acid at a low pH appears to depend on their H⁺-exporting ability, but this requires not only a highly active H⁺-ATPase in the plasma membrane but the provision of sufficient ATP to drive this energy-demanding process as indicated by the results of Holyoak et al. (8). It is likely that increased ATPase activity under octanoic acid stress may reduce cellular ATP levels and that ATP depletion contributes to the failure of the maintenance of pH_i homeostasis, particularly among the subpopulation that in the inoculum exhibited the lowest pH_i values. The loss of viability might occur for those cells where pH_i decreased down to nonphysiological values. The eventual recovery of growth therefore depends on the remaining viable population, in agreement with the well-known critical role played by the physiology and number of viable cells in the inoculum in the duration of latency under acid stress. The observation that octanoic acid-adapted cells reinoculated into the same fresh medium can resume growth after a much shorter latency (Fig. ​(Fig.1a)1a) is a good example of the importance of the physiology of the inoculum cells. Besides the increased plasma membrane H⁺-ATPase activity of octanoic acid-adapted cells, other mechanisms may underlie the adaptation to acid stress, such as the increased cellular buffering capacity of octanoic acid-grown cells due to their lower intracellular volume (20), the more favorable plasma membrane lipid composition (1), and the possible induction of the active efflux of the anion (26).     

Mutational Biosynthesis of Novel Rapamycins by a Strain of Streptomyces hygroscopicus NRRL 5491 Disrupted in rapL,Encoding a Putative Lysine Cyclodeaminase

The gene rapL lies within the region of the Streptomyces hygroscopicus chromosome which contains the biosynthetic gene cluster for the immunosuppressant rapamycin. Introduction of a frameshift mutation into rapL by ΦC31 phage-mediated gene replacement gave rise to a mutant which did not produce significant amounts of rapamycin. Growth of this rapL mutant on media containing added l-pipecolate restored wild-type levels of rapamycin production, consistent with a proposal that rapL encodes a specific l-lysine cyclodeaminase important for the production of the l-pipecolate precursor. In the presence of added proline derivatives, rapL mutants synthesized novel rapamycin analogs, indicating a relaxed substrate specificity for the enzyme catalyzing pipecolate incorporation into the macrocycle.Rapamycin is a 31-member macrocyclic polyketide produced by Streptomyces hygroscopicus NRRL 5491 which, like the structurally related compounds FK506 and immunomycin (Fig. ​(Fig.1),1), has potent immunosuppressive properties (24). Such compounds are potentially valuable in the treatment of autoimmune diseases and in preventing the rejection of transplanted tissues (16). The biosynthesis of rapamycin requires a modular polyketide synthase, which uses a shikimate-derived starter unit (11, 20) and which carries out a total of fourteen successive cycles of polyketide chain elongation that resemble the steps in fatty acid biosynthesis (2, 27). l-Pipecolic acid is then incorporated (21) into the chain, followed by closure of the macrocyclic ring, and both these steps are believed to be catalyzed by a pipecolate-incorporating enzyme (PIE) (18), the product of the rapP gene (8, 15). Further site-specific oxidations and O-methylation steps (15) are then required to produce rapamycin. Open in a separate windowFIG. 1Structures of rapamycin, FK506, and immunomycin.The origin of the pipecolic acid inserted into rapamycin has been previously established (21) to be free l-pipecolic acid derived from l-lysine (although the possible role of d-lysine as a precursor must also be borne in mind) (9). Previous work with other systems has suggested several alternative pathways for pipecolate formation from lysine (22), but the results of the incorporation of labelled lysine into the pipecolate moiety of immunomycin (Fig. ​(Fig.1)1) clearly indicate loss of the α-nitrogen atom (3). More recently, the sequencing of the rap gene cluster revealed the presence of the rapL gene (Fig. ​(Fig.2),2), whose deduced gene product bears striking sequence similarity to two isoenzymes of ornithine deaminase from Agrobacterium tumefaciens (25, 26). Ornithine deaminase catalyzes the deaminative cyclization of ornithine to proline, and we have proposed (15) that the rapL gene product catalyzes the analogous conversion of l-lysine to l-pipecolate (Fig. ​(Fig.3).3). Open in a separate windowFIG. 2A portion of the rapamycin biosynthetic gene cluster which contains ancillary (non-polyketide synthase) genes (15, 27). PKS, polyketide synthase.Open in a separate windowFIG. 3(A) The conversion of l-ornithine to l-proline by ornithine cyclodeaminase (17). (B) Proposed conversion of l-lysine to l-pipecolic acid by the rapL gene product.Here, we report the use of ΦC31 phage-mediated gene replacement (10) to introduce a frameshift mutation into rapL and the ability of the mutant to synthesize rapamycins in the absence or presence of added pipecolate or pipecolate analogs.     

Dissecting an Allosteric Switch in Caspase-7 Using Chemical and Mutational Probes

Apoptotic caspases, such as caspase-7, are stored as inactive protease zymogens, and when activated, lead to a fate-determining switch to induce cell death. We previously discovered small molecule thiol-containing inhibitors that when tethered revealed an allosteric site and trapped a conformation similar to the zymogen form of the enzyme. We noted three structural transitions that the compounds induced: (i) breaking of an interaction between Tyr-223 and Arg-187 in the allosteric site, which prevents proper ordering of the catalytic cysteine; (ii) pinning the L2′ loop over the allosteric site, which blocks critical interactions for proper ordering of the substrate-binding groove; and (iii) a hinge-like rotation at Gly-188 positioned after the catalytic Cys-186 and Arg-187. Here we report a systematic mutational analysis of these regions to dissect their functional importance to mediate the allosteric transition induced by these compounds. Mutating the hinge Gly-188 to the restrictive proline causes a massive ∼6000-fold reduction in catalytic efficiency. Mutations in the Arg-187–Tyr-223 couple have a far less dramatic effect (3–20-fold reductions). Interestingly, although the allosteric couple mutants still allow binding and allosteric inhibition, they partially relieve the mutual exclusivity of binding between inhibitors at the active and allosteric sites. These data highlight a small set of residues critical for mediating the transition from active to inactive zymogen-like states.Caspases are a family of dimeric cysteine proteases whose members control the ultimate steps for apoptosis (programmed cell death) or innate inflammation among others (for reviews, see Refs. 1 and 2). During apoptosis, the upstream initiator caspases (caspase-8 and -9) activate the downstream executioner caspases (caspase-3, -6, and-7) via zymogen maturation (3). The activated executioner caspases then cleave upwards of 500 key proteins (4–6) and DNA, leading to cell death. Due to their pivotal role in apoptosis, the caspases are involved both in embryonic development and in dysfunction in diseases including cancer and stroke (7). The 11 human caspases share a common active site cysteine-histidine dyad (8), and derive their name, cysteine aspartate proteases, from their exquisite specificity for cleaving substrate proteins after specific aspartate residues (9–13). Thus, it has been difficult to develop active site-directed inhibitors with significant specificity for one caspase over the others (14). Despite difficulties in obtaining specificity, there has been a long-standing correlation between efficacy of caspase inhibitors in vitro and their ability to inhibit caspases and apoptosis in vivo (for review, see Ref. 31). Thus, a clear understanding of in vitro inhibitor function is central to the ability control caspase function in vivo.Caspase-7 has been a paradigm for understanding the structure and dynamics of the executioner caspases (15–21). The substrate-binding site is composed of four loops; L2, L3, and L4 are contributed from one-half of the caspase dimer, and L2′ is contributed from the other half of the caspase dimer (Fig. 1). These loops appear highly dynamic as they are only observed in x-ray structures when bound to substrate or substrate analogs in the catalytically competent conformation (17–19, 22) (Fig. 1B).Open in a separate windowFIGURE 1.Allosteric site and dimeric structure in caspase-7. A, the surface of active site-bound caspase-7 shows a large open allosteric (yellow) site at the dimer interface. This cavity is distinct from the active sites, which are bound with the active site inhibitor DEVD (green sticks). B, large subunits of caspase-7 dimers (dark green and dark purple) contain the active site cysteine-histidine dyad. The small subunits (light green and light purple) contain the allosteric site cysteine 290. The conformation of the substrate-binding loops (L2, L2′, L3, and L4) in active caspase-7 (Protein Data Bank (PDB) number 1f1j) is depicted. The L2′ loop (spheres) from one-half of the dimer interacts with the L2 loop from the other half of the dimer. C, binding of allosteric inhibitors influences the conformation of the L2′ loop (spheres), which folds over the allosteric cavity (PDB number 1shj). Subunit rendering is as in panel A. Panels A, B, and C are in the same orientation.A potential alternative to active site inhibitors are allosteric inhibitors that have been seeded by the discovery of selective cysteine-tethered allosteric inhibitors for either apoptotic executioner caspase-3 or apoptotic executioner caspase-7 (23) as well as the inflammatory caspase-1 (24). These thiol-containing compounds bind to a putative allosteric site through disulfide bond formation with a thiol in the cavity at the dimer interface (Fig. 1A) (23, 24). X-ray structures of caspase-7 bound to allosteric inhibitors FICA³ and DICA (Fig. 2) show that these compounds trigger conformational rearrangements that stabilize the inactive zymogen-like conformation over the substrate-bound, active conformation. The ability of small molecules to hold mature caspase-7 in a conformation that mimics the naturally occurring, inactive zymogen state underscores the utility and biological relevance of the allosteric mechanism of inhibition. Several structural changes are evident between these allosterically inhibited and active states. (i) The allosteric inhibitors directly disrupt an interaction between Arg-187 (next to the catalytic Cys-186) and Tyr-223 that springs the Arg-187 into the active site (Fig. 3), (ii) this conformational change appears to be facilitated by a hinge-like movement about Gly-188, and (iii) the L2′ loop folds down to cover the allosteric inhibitor and assumes a zymogen-like conformation (Fig. 1C) (23).Open in a separate windowFIGURE 2.Structure of allosteric inhibitors DICA and FICA. DICA and FICA are hydrophobic small molecules that bind to an allosteric site at the dimer interface of caspase-7. Binding of DICA/FICA is mediated by a disulfide between the compound thiol and Cys-290 in caspase-7.Open in a separate windowFIGURE 3.Movement of L2′ blocking arm. The region of caspase-7 encompassing the allosteric couple Arg-187 and Tyr-223 is boxed. The inset shows the down orientation of Arg-187 and Tyr-223 in the active conformation with DEVD substrate mimic (orange spheres) in the active site. In the allosteric/zymogen conformation, Arg-187 and Tyr-223 are pushed up by DICA (blue spheres).Here, using mutational analysis and small molecule inhibitors, we assess the importance of these three structural units to modulate both the inhibition of the enzyme and the coupling between allosteric and active site labeling. Our data suggest that the hinge movement and pinning of the L2-L2′ are most critical for transitioning between the active and inactive forms of the enzyme.     

Functionality and Cell Anchorage Dependence of the African Swine Fever Virus Gene A179L,a Viral bcl-2 Homolog,in Insect Cells

The African swine fever virus gene A179L has been shown to be a functional member of the ced9/bcl-2 family of apoptosis inhibitors in mammalian cell lines. In this work we have expressed the A179L gene product (p21) under the control of the baculovirus polyhedrin promoter using a baculovirus system. Expression of the A179L gene neither altered the baculovirus replication phenotype nor delayed the shutoff of cellular protein synthesis, but it extended the survival of the infected insect cells to very late times postinfection. The increase in cell survival rates correlated with a marked apoptosis reduction after baculovirus infection. Interestingly, prevention of apoptosis was observed when recombinant baculovirus infections were carried out in monolayer cell cultures but not when cells were infected in suspension, suggesting a cell anchorage dependence for p21 function in insect cells. Cell survival was enhanced under optimal conditions of cell attachment and cell-to-cell contact as provided by extracellular matrix components or poly-d-lysine. Since it was observed that cytoskeleton organization varied depending on culture conditions of insect cells (grown in monolayer versus grown in suspension), these results suggested that A179L might regulate apoptosis in insect cells only when the cytoskeletal support of intracellular signaling is maintained upon cell adhesion. Thus, cell shape and cytoskeleton status might allow variations in intracellular transduction of signals related to cell survival in virus-infected cells.African swine fever virus (ASFV), the causative agent for an important disease of swine, is a large double-stranded DNA virus that replicates not only in members of the Suidae family but also in soft ticks of the Ornithodoros genus (24, 31, 41). ASFV infects a variety of cells of the mononuclear phagocytic system and produces a characteristic apoptotic cell death in infected swine macrophages and bystander uninfected lymphocytes (33–35). Virus-induced apoptosis in target cells is produced late during infection, suggesting the existence of viral genes that prevent early apoptosis to support the productive infection. An ASFV gene, 5HL (A179L in Ba71V virus), with sequence similarity to the ced9/bcl-2 gene family has been described (28). This gene encodes a protein of approximately 21 kDa (p21) that is synthesized in infected cells at both early and late times postinfection (28). The similarity of this gene to bcl-2 has pointed to its possible role in apoptosis inhibition during ASFV infection. We have recently demonstrated that expression of the A179L gene by a recombinant vaccinia virus inhibits apoptosis mediated by the interferon-induced double-stranded RNA-activated protein kinase in HeLa and BSC-40 cells (5). Also, the A179L product is able to suppress apoptotic cell death in the FL5.12 mouse prolymphocytic cell line and in the K562 human myeloid leukemia cell line (1, 37). Thus, the ability of the A179L gene to suppress apoptosis in mammalian cells has been clearly shown. However, the function in insect cells of the bcl-2 gene, the prototype of a death-regulator gene family, remains controversial (3, 8, 11), and the function of A179L during ASFV infection of nonmammalian cells is still unknown. Other ASFV genes potentially involved in the regulation of intracellular apoptosis pathways are A238L and 23NL, which have sequence similarity to the IκB factor (32) and the ICP34.5 gene of herpes simplex virus I (14, 40), respectively. The existence of an ASFV gene (A224L) with similarity to the iap family of apoptosis inhibitors suggested that A179L and A224L could function as host range genes (9). Nevertheless, there are no current data supporting the function of A224L as an apoptosis inhibitor, and studies conducted to analyze its role during ASFV infection demonstrated that this gene is dispensable for ASFV growth in swine macrophages (29).Therefore, the study of the apoptosis regulatory functions of the A179L gene in different cell lines and under different culture conditions is important to an understanding of the role of this gene during ASFV infection in its different hosts. To examine these functions, we have expressed the A179L gene in Spodoptera frugiperda (Sf9) cells using a baculovirus system. We demonstrated that this gene is functional and prevents virus-induced apoptosis in these cells only when intracellular signaling upon cell adhesion is maintained.

Expression of the A179L gene product in Sf9 cells.

The construction of recombinant baculovirus expressing the A179L gene product was carried out as previously described (21, 30). Briefly, DNA amplification of the A179L gene from the ASFV isolate E70 was carried out by PCR with ampliTaq DNA polymerase (Perkin-Elmer Cetus) with the primers (i) 5′-AAATATAGGGATCCGCTATGGAGGG (5′ primer) and (ii) 5′-CCGCGTGGATCCTATATCAAATTGC (3′ primer). Both primers contain the recognition sequence for the BamHI restriction enzyme. The PCR product was digested with BamHI and cloned into the BamHI site of the baculovirus transfer vector pBacPAK8 (Clontech) under the control of the polyhedrin promoter. The cloned gene was sequenced by the dideoxynucleotide chain terminator method by using specific primers to check possible sequence changes introduced by PCR amplification of the gene. Then, Sf9 cells were cotransfected with the transfer recombinant vector (pBacPAK-A179L) and the purified, noninfectious BacPAK6 DNA (Bsu36I digested), which contains the β-galactosidase gene under the control of the polyhedrin promoter. Isolation of recombinant baculovirus was achieved by negative selection due to replacement of the β-galactosidase gene by the newly introduced recombinant gene. The selected virus was further purified after three successive plaque assays in Sf9 cells. The recombinant baculovirus expressing the A179L gene was denominated BVA179L. Another recombinant baculovirus bearing the A179L gene cloned in the opposite orientation (antisense, BVA179L-AS) was constructed by similar methods.Expression of the A179L gene by the recombinant baculovirus BVA179L was analyzed by Western blotting at 72 h postinfection (p.i.) in cells from suspension and monolayer cultures (Fig. ​(Fig.1A).1A). Cell extracts with similar numbers of infected cells from both cultures were lysed, electrophoresed in sodium dodecyl sulfate (SDS)–15% polyacrylamide gels, transferred to nitrocellulose filters (Bio-Rad), and then incubated with a polyclonal antiserum raised against the product of the ASFV gene 5HL expressed in Escherichia coli cells (28). The serum reacted with a 21-kDa polypeptide (p21) but failed to react with any protein product of this electrophoretic mobility in mock-infected cells (data not shown) or in cells infected with the recombinant baculovirus BVA179L-AS (Fig. ​(Fig.1A).1A). Open in a separate windowFIG. 1Expression of the ASFV A179L gene by using a baculovirus system. (A) Western blot analysis of infected Sf9 cell extracts. Cell lysates from Sf9 cells synchronously infected either with the recombinant baculovirus BVA179L or with BVA179L-AS reacted at 72 h p.i. with a specific antiserum raised against the A179L gene product expressed in E. coli cells. (B) Immunofluorescence of BVA179L- or BVA179L-AS-infected insect cells in suspension, analyzed at 72 h p.i. by using the same serum against A179L stained with fluorescein isothiocyanate (FITC). Green cytoplasmic fluorescence was detected only in cells infected with BVA179L. The cell nucleus was contrasted with propidium iodide (PI) (right). (C) Growth curves of BVA179L and BVA179L-AS recombinant baculoviruses in Sf9 monolayer cell cultures. Extracellular virus obtained from culture supernatants at the indicated time points was titrated in a conventional plaque assay. Squares, BVA179L-AS; circles, BVA179L.Indirect immunofluorescence of fixed and permeabilized BVA179L- or BVA179L-AS-infected Sf9 cells, using the same anti-p21 serum, revealed predominantly cytoplasmic staining for p21 (Fig. ​(Fig.1B,1B, center), which contrasted with the nuclear localization of propidium iodide, a DNA-intercalating agent (Fig. ​(Fig.1B,1B, right). This distribution of p21 was similar in cells grown in suspension or in adherent growth conditions.The effect of p21 expression on recombinant baculovirus growth was also investigated (Fig. ​(Fig.1C).1C). Cells were infected (multiplicity of infection [MOI], 2), supernatants were collected at different times postinfection for titration, and fresh medium was added to the cultures. Infection with BVA179L or BVA179L-AS yielded similar virus titers, suggesting that overexpression of the A179L gene did not alter the baculovirus replication phenotype in Sf9 cells. Maximum viral titers were reached at 48 h p.i. with both recombinant viruses (Fig. ​(Fig.1C).1C). As expected, virus yields dropped drastically after this time point, because the experiment measured virus release to the extracellular medium rather than virus accumulation. This result suggested the occurrence of a strong shutoff of protein synthesis due to baculovirus infection at late time points.In order to confirm this fact, long-term synthesis of baculovirus-induced proteins in the presence of the apoptosis inhibitor p21 was analyzed by metabolic pulse labeling of infected cells at different times postinfection. Sf9 cells (10⁵) were mock infected or infected at a MOI of 10 in 96-well culture plates. At 22, 46, 70, and 94 h p.i. the medium was replaced with Grace’s medium lacking methionine (Gibco) and maintained for 1 h prior to the addition of fresh methionine-deficient medium containing 200 μCi of [³⁵S]methionine/ml (>1,000 Ci/mmol). At the selected time points cell pellets were harvested, lysed in SDS buffer, and analyzed by autoradiography after SDS-polyacrylamide gel electrophoresis. A strong shutoff of protein synthesis was observed starting from 48 h p.i. in monolayer cultures infected with BVA179L, BVA179L-AS, or a baculovirus expressing β-galactosidase (data not shown). At 96 h p.i. no newly synthesized cellular or viral-induced proteins were detected in any infected culture.

Functionality of the A179L gene in insect cells.

To determine the possible role of the A179L gene in prevention of baculovirus-induced apoptosis, viability of cell cultures infected at MOI of 2, 10 and 100 was determined at various times postinfection by trypan blue exclusion by counting 1 × 10³ to 1.4 × 10³ cells in five independent fields each (Fig. ​(Fig.2).2). High MOI were used to assure a synchronized infection with 99% infected cells. Viral infections on monolayer cell cultures were carried out in multiwell plastic dishes (Nunc). For the infection of suspension cultures, 250-ml flasks in a rotary shaker, with constant stirring at 80 rpm, were used. Both monolayer and suspension cultures were maintained in Grace’s insect medium (Gibco-BRL) supplemented with 10% fetal bovine serum. Cells were inoculated with extracellular, budded virus, and viral titers were determined by plaque assay on 10⁶ cells seeded onto 35-mm dishes and overlaid with a mixture of 0.7% agarose (Sigma) in 10% fetal bovine serum–Grace’s medium. Infections with a control baculovirus expressing the reporter gene β-galactosidase at a MOI of 10 or higher yielded more than 95% cells expressing the reporter gene in both monolayer and suspension (data not shown). With all recombinants, all cells showed a clear cytopathic effect characteristic of baculovirus productive infection. Open in a separate windowFIG. 2Effect of A179L gene expression on the survival of baculovirus-infected Sf9 cells. Viability of insect cells grown in monolayers (A) or in suspension (B) and infected with BVA179L-AS (squares) or BVA179L (circles) at a MOI of 100 was determined by trypan blue exclusion at different times postinfection. Means and standard errors were calculated from three independent experiments. (C) A representative field of trypan blue-stained monolayer cultures of BVA179L- or BVA179L-AS-infected Sf9 cells at 144 h p.i. Original magnification, ×200.The viability levels of baculovirus-infected cells correlated with the postinfection time. Differences in cell viability were consistently found after 48 h p.i., when viral yields reached a maximum and protein shutoff was more evident. In monolayer cultures (Fig. ​(Fig.2A),2A), fewer than 20% of the BVA179L-AS-infected cells survived to infection at 144 h p.i. (Fig. ​(Fig.2A).2A). In contrast, cells infected with the baculovirus expressing the A179L gene presented only a slight decrease in viability at this time point (Fig. ​(Fig.2A).2A). A representative field of cells infected with recombinant baculoviruses stained with trypan blue at 144 h p.i. is shown in Fig. ​Fig.2C.2C. Thus, the expression of p21 at late times postinfection increased the survival of baculovirus-infected Sf9 cells cultured in monolayer.However, in suspension, viability experiments carried out with cells infected with either BVA179L or BVA179L-AS did not result in demonstrable differences in survival rates (Fig. ​(Fig.2B).2B). High proportions of dead cells (about 50%) were found at 48 h p.i. in spite of p21 expression. These discrepancies found between monolayer and suspension cultures suggested that the effect of p21 expression on Sf9 viability could be related to the lack of cell attachment to a substrate in suspended cultures.We then investigated whether the increased viability found in BVA179L-infected cells in monolayer cultures was due to apoptosis inhibition. Since both viruses expressed identical levels of baculovirus p35 at early times (11), infection-induced apoptosis should be similar in both cases, thus minimizing the differences in survival rates before 48 h p.i. Beyond this time point, when p35 is no longer functional (8, 11), differences between BVA179L and BVA179L-AS, not expressing p21, might become evident. As programmed cell death is relatively highly conserved during evolution and inhibitors of apoptosis are functionally interchangeable among distant species, it might be reasonable to suggest a function for A179L in insect cells similar to that displayed in mammalian cells. However, the bcl-2 homolog gene function in insect cells still remains controversial. Alnemri et al. (3) found that overexpression of human bcl-2 increased survival of baculovirus-infected Sf9 cells by prevention of apoptosis. Since the gene encoding p21 is a bcl-2 homolog (1, 5, 37), it seems likely that both genes act in similar apoptosis pathways. Nevertheless, it was reported that expression of bcl-2 or the adenovirus gene E1B-19K did not rescue the wild-type phenotypes of baculoviruses lacking the p35 gene (8, 11), so it was postulated that early apoptosis induction prevented by p35 expression could be mediated by bcl-2-independent mechanisms in Sf9 cells. Baculovirus infection could then trigger two different apoptosis mechanisms, an early apoptosis blocked by p35 but not by bcl-2 or its homologs, such as p21, and a late apoptosis induction in which p21/bcl-2 might be functional.The chromatin fragmentation of baculovirus-infected cells by different methods was then analyzed (Fig. ​(Fig.3A3A to C). First we carried out a comparative Hoechst 33258 staining of infected cells in monolayer at different hours postinfection with BVA179L or BVA179L-AS viruses (Fig. ​(Fig.3A).3A). Cells were methanol fixed for 10 min before incubation with 10 μg of Hoechst dye per ml in phosphate-buffered saline for 30 min at room temperature. BVA179L-AS-virus-infected cells exhibited a chromatin fragmentation pattern characteristic of apoptosis (Fig. ​(Fig.3A,3A, right) which increased in a time-dependent fashion (data not shown). In contrast, at 144 h p.i. Sf9 cells infected with the BVA179L virus presented very few figures of apoptosis (Fig. ​(Fig.3A,3A, left). This result indicated a correlation between cell viability and occurrence of chromatin fragmentation, which was confirmed by DNA laddering analysis (Fig. ​(Fig.3C)3C) in 1.6% agarose gels (5). These experiments confirmed the lack of apoptosis prevention by p21 under nonadherent cell culture growth conditions (Fig. ​(Fig.3C,3C, right). Open in a separate windowFIG. 3DNA fragmentation in infected Sf9 cells (MOI of 10). (A) Hoechst 33258 staining of BVA179L- or BVA179L-AS-infected Sf9 cells grown in plates at 144 h p.i. Original magnification, ×400. (B) ELISA quantitation of the DNA linked to histone proteins in the cytoplasmic fraction of apoptotic Sf9 cells infected in monolayer at different time points. Squares: BVA179L-AS-infected cells; circles, BVA179L-infected cells; triangles, mock-infected cells. Means and standard errors were calculated from three independent experiments. (C) Agarose gel electrophoresis of the internucleosomal DNA laddering detected in infected or mock-infected Sf9 cells in monolayer cultures at 96 and 144 h p.i. (left) or in suspension cultures at 96 h p.i. (right). M, molecular size markers.Quantitation of histone-associated DNA fragments released to the cytoplasm was carried out by a specific enzyme-linked immunosorbent assay (ELISA) (Boehringer Mannheim) (5). The results obtained after this analysis clearly indicated that expression of the ASFV gene A179L prevented the onset of apoptosis induced by baculovirus infection of cells cultured in monolayer (Fig. ​(Fig.3B).3B). In contrast, at 144 h p.i. BVA179L-AS-virus-infected cells (Fig. ​(Fig.3B)3B) yielded higher apoptosis rates than cells expressing p21.The above results strongly suggest that the antiapoptotic function of p21 is dependent on cell attachment, because in suspended cells, A179L expression was unable to prevent baculovirus-induced apoptosis. Cell attachment is important for many cell functions. In fact, most types of normal cells require extracellular matrix attachment to respond to growth factor stimulation and other signals controlling cell proliferation or survival. When detached from their matrix, some cells undergo apoptosis (7, 15, 19, 26).

Effect of cell attachment on p21 function.

In an attempt to confirm if the activity of p21 was dependent on cell attachment, we performed experiments by infecting Sf9 cells with recombinant baculoviruses under conditions that improved cell adhesion to the culture surface (Fig. ​(Fig.4A).4A). Sf9 cells were cultured on extracellular matrix components, such as rat collagen type I and human fibronectin, or on poly-d-lysine-coated glass plates (Falcon). Then, cells were infected with BVA179L or BVA179L-AS viruses at a MOI of 10. Figure ​Figure4A4A shows the apoptosis indices (AIs) of those cultures measured by anti-histone quantitative ELISA (inverse correlation). In BVA179L-infected cultures, apoptosis rate reduction was detected with all substrates that facilitated cell attachment by any pathway. Either extracellular matrix components (collagen type I and fibronectin) that support integrin-mediated cell adhesion or a substrate that mediates adhesion by nonspecific interactions (poly-d-lysine) enhanced apoptosis protection with respect to apoptosis of cells grown on uncoated nonadherent glass surfaces (AI of ≤1). Interestingly, infections performed on cells on a collagen type I matrix yielded more differences in apoptosis inhibition by p21 (AI of ≤0.5) with respect to apoptosis of cells on untreated surfaces (Fig. ​(Fig.4A).4A). This could be related to the fact that Sf9 cells grown on this type of surface also had increased cell-to-cell interactions (not shown). Open in a separate windowFIG. 4(A) Apoptosis induction of Sf9 cells (expressing p21) expressed as the ratio to that in BVA179L-AS-infected cells grown on different substrates to improve adhesiveness, measured by ELISA. The AI is given by the quotient (BVA179L-infected cells/BVA179L-AS-infected cells) of mean absorbance values at 405 nm from two replicate experiments at 96 h p.i. FN, human fibronectin; PDL, poly-d-lysine; COLI, rat tail collagen type I; UNTREATED, BVA179L-infected cells. (B) Actin cytoskeleton organization in adherent insect cells (left). Cells cultured in suspended growth conditions presented actin focalization (right). Cellular F-actin was stained with phalloidin-tetramethyl rhodamine isothiocyanate. (C) Characteristic individual infected Sf9 cells cultured in monolayer (upper) or in suspension (lower). Actin clumped in coarse fragments showed intense red staining. Original magnification, ×600 (panels B and C).Cell attachment to underlying extracellular matrix is mediated by specialized membrane proteins called integrins, which interact with determined extracellular matrix components (6, 22). Integrin-mediated cell adhesion initiates a cascade of events that allow the transduction of survival signals that can block programmed cell death (13, 25). Induction of this survival pathway includes the upregulation of antiapoptotic proteins such as bcl-2 family members (16, 27, 42, 43). Also, cell-to-cell interaction inversely correlates with apoptosis associated with bcl-2 protein expression (4); in our results, aggregation of Sf9 cells plated on collagen I matrix resulted in enhanced survival. Apoptosis inhibition by p21 was also obtained with a nonspecific adhesive, such as poly-d-lysine, a synthetic compound altering surface charge, that increases cell adhesion not mediated by integrins. Our results indicate that cell attachment alone is sufficient to allow for the antiapoptotic activity of p21 in Sf9 cells infected by baculovirus. In fact, recent findings focus on cell shape changes and cytoskeleton integrity as supportive of rescue from apoptosis (10, 38).

Cytoskeleton organization in insect cells during infection with baculovirus in presence of p21.

Since conditions of cell anchorage may modify cytoskeleton organization, we analyzed the actin cytoskeleton of insect cells cultured under adherent and suspended growth conditions (Fig. ​(Fig.4).4). Sf9 cells were grown either directly on 96-well multiwell plates or in spinner flasks. Cells from suspension cultures were removed and allowed to sediment on glass slides. In both cases, cells were fixed with 4% paraformaldehyde, permeabilized in 0.1% Triton X-100 in phosphate-buffered saline, and stained for 30 min with 1:300 phalloidin-tetramethyl rhodamine isothiocyanate (Sigma), a marker for F-actin. Uninfected Sf9 cells attached to a surface displayed morphology different from that of cells that were grown in suspension. A profuse and fine surface microvillar network present in attached cells (Fig. ​(Fig.4B,4B, left) was lacking in suspension. In contrast, suspended cells showed marked focalization of actin staining, indicating the occurrence of cytoskeleton reorganization (Fig. ​(Fig.4B,4B, right). Cells in suspension infected with either BVA179L or BVA179L-AS baculoviruses showed irregular clumping of actin (Fig. ​(Fig.4C,4C, lower). Dense actin staining was found concentrated in coarse fragments, and such elements were found in the cultures in proportions similar to that in the nonviable cell fraction (Fig. ​(Fig.2A2A and B). However, this pattern of actin clumping was not found in attached cells infected with BVA179L (Fig. ​(Fig.4C,4C, upper left). Moreover, in monolayer, the overall intensity of actin staining decreased in a time-dependent manner in cells infected with BVA179L-AS, but it was maintained longer in infected cells expressing p21 (not shown). This result suggests that the expression of p21 in BVA179L-infected cells could contribute to the preservation of the cellular actin cytoskeleton at late postinfection times.Cell adhesion to underlying extracellular matrix is mediated by sites of tight adhesion, called focal adhesions, that develop in cells in culture (12). Focal adhesions provide a structural link between the actin cytoskeleton and the extracellular matrix and are regions of signal transduction related to gene expression, growth control, and cell survival. It was recently suggested that cell attachment to matrix or integrin binding per se is not sufficient for maintaining cell viability and that cells need to undergo some minimal degree of shape changes to survive (10, 36, 38). It was recently reported that suspended endothelial cells acquired rounded shape, presented cytoskeleton disorganization, and underwent apoptosis (36). In contrast, when cells were grown on fibronectin or vitronectin, they became flattened, showed actin microfilament organization, and retained viability (36). Interleukin 4, which is able to activate neutrophil cytoskeletal rearrangements, produces a delay of apoptosis (20). Also, epithelial cells cultured on extracellular matrix components or laminin had a more well-developed actin cytoskeleton than cells cultured on noncoated dishes, which underwent apoptosis (2). The organization of the actin cytoskeleton in Sf9 cells grown attached to a surface was quite different from that displayed by cells grown in suspension. Consequently, actin organization and cell shape changes might provide the conditions for p21 protective function.The in vivo relevance of changes in cell anchorage has been mainly focused on to date, either in the leukocyte movement out of vessels along endothelial cells in inflammation (23, 39) or in the loss of adherence of transformed cells in metastasis production (17, 18). Our findings suggest a role for cell shape and cytoskeleton status in viral diseases as well. Variations in those conditions in determined tissues or cell types in viral infections might explain differences in intracellular transduction of signals related to cell growth and survival. The cell anchorage dependence demonstrated by the ASFV bcl-2 homolog could have important consequences in the infection among the different cell compartments in vivo. Nevertheless, the physiological relevance of the biological effects of a protein overexpressed to the levels reported here should be an object of future studies. Based on these findings, it should be further analyzed if the absence of function of A179L apoptosis inhibitor in nonattached infected cells, such as circulating cells, might favor an early apoptosis induction and death of those cells. In fact, during in vivo infection with ASFV, only small percentages of infected monocytes are detected in peripheral blood (33). In contrast, tissue-fixed macrophages attached to extracellular matrix could be more prone to support the function of the bcl-2 viral homolog in vivo, leading to apoptosis inhibition in these cells, and could constitute a viral reservoir during persistent infection.In conclusion, the above-presented data demonstrate that the product of the ASFV A179L gene, p21, is functional in insect cells and prevents late apoptosis after baculovirus infection. p21 increases the viability of infected cells in the context of a strong shutoff of protein synthesis and without modifying the baculovirus infection cycle. This suggests that p21 probably inhibits, in a way similar to human bcl-2, a highly conserved component of the apoptosis execution program.