Experimental Evolution Studies in Φ6 Cystovirus.

Experimental evolution studies, in which biological populations are evolved in a specific environment over time, can address questions about the nature of spontaneous mutations, responses to selection, and the origins and maintenance of novel traits. Here, we review more than 30 years of experimental evolution studies using the bacteriophage (phage) Φ6 cystovirus. Similar to many lab-studied bacteriophages, Φ6 has a high mutation rate, large population size, fast generation time, and can be genetically engineered or cryogenically frozen, which facilitates its rapid evolution in the laboratory and the subsequent characterization of the effects of its mutations. Moreover, its segmented RNA genome, outer membrane, and capacity for multiple phages to coinfect a single host cell make Φ6 a good non-pathogenic model for investigating the evolution of RNA viruses that infect humans. We describe experiments that used Φ6 to address the fitness effects of spontaneous mutations, the consequences of evolution in the presence of coinfection, the evolution of host ranges, and mechanisms and consequences of the evolution of thermostability. We highlight open areas of inquiry where further experimentation on Φ6 could inform predictions for pathogenic viruses.

Next-generation sequencing of dsRNA is greatly improved by treatment with the inexpensive denaturing reagent DMSO.

dsRNA is the genetic material of important viruses and a key component of RNA interference-based immunity in eukaryotes. Previous studies have noted difficulties in determining the sequence of dsRNA molecules that have affected studies of immune function and estimates of viral diversity in nature. DMSO has been used to denature dsRNA prior to the reverse-transcription stage to improve reverse transcriptase PCR and Sanger sequencing. We systematically tested the utility of DMSO to improve the sequencing yield of a dsRNA virus (Φ6) in a short-read next-generation sequencing platform. DMSO treatment improved sequencing read recovery by over two orders of magnitude, even when RNA and cDNA concentrations were below the limit of detection. We also tested the effects of DMSO on a mock eukaryotic viral community and found that dsRNA virus reads increased with DMSO treatment. Furthermore, we provide evidence that DMSO treatment does not adversely affect recovery of reads from a ssRNA viral genome (influenza A/California/07/2009). We suggest that up to 50 % DMSO treatment be used prior to cDNA synthesis when samples of interest are composed of or may contain dsRNA.

Multiple liquid crystalline geometries of highly compacted nucleic acid in a dsRNA virus.

Characterizing the genome of mature virions is pivotal to understanding the highly dynamic processes of virus assembly and infection. Owing to the different cellular fates of DNA and RNA, the life cycles of double-stranded (ds)DNA and dsRNA viruses are dissimilar. In terms of nucleic acid packing, dsDNA viruses, which lack genome segmentation and intra-capsid transcriptional machinery, predominantly display single-spooled genome organizations1-8. Because the release of dsRNA into the cytoplasm triggers host defence mechanisms9, dsRNA viruses retain their genomes within a core particle that contains the enzymes required for RNA replication and transcription10-12. The genomes of dsRNA viruses vary greatly in the degree of segmentation. In members of the Reoviridae family, genomes consist of 10-12 segments and exhibit a non-spooled arrangement mediated by RNA-dependent RNA polymerases11-14. However, whether this arrangement is a general feature of dsRNA viruses remains unknown. Here, using cryo-electron microscopy to resolve the dsRNA genome structure of the tri-segmented bacteriophage ɸ6 of the Cystoviridae family, we show that dsRNA viruses can adopt a dsDNA-like single-spooled genome organization. We find that in this group of viruses, RNA-dependent RNA polymerases do not direct genome ordering, and the dsRNA can adopt multiple conformations. We build a model that encompasses 90% of the genome, and use this to quantify variation in the packing density and to characterize the different liquid crystalline geometries that are exhibited by the tightly compacted nucleic acid. Our results demonstrate that the canonical model for the packing of dsDNA can be extended to dsRNA viruses.

Existing Host Range Mutations Constrain Further Emergence of RNA Viruses.

RNA viruses are capable of rapid host shifting, typically due to a point mutation that confers expanded host range. As additional point mutations are necessary for further expansions, epistasis among host range mutations can potentially affect the mutational neighborhood and frequency of niche expansion. We mapped the mutational neighborhood of host range expansion using three genotypes of the double-stranded RNA (dsRNA) bacteriophage φ6 (wild type and two isogenic host range mutants) on the novel host Pseudomonas syringae pv. atrofaciens. Both Sanger sequencing of 50 P. syringae pv. atrofaciens mutant clones for each genotype and population Illumina sequencing revealed the same high-frequency mutations allowing infection of P. syringae pv. atrofaciens. Wild-type φ6 had at least nine different ways of mutating to enter the novel host, eight of which are in p3 (host attachment protein gene), and 13/50 clones had unchanged p3 genes. However, the two isogenic mutants had dramatically restricted neighborhoods: only one or two mutations, all in p3. Deep sequencing revealed that wild-type clones without mutations in p3 likely had changes in p12 (morphogenic protein), a region that was not polymorphic for the two isogenic host range mutants. Sanger sequencing confirmed that 10/13 of the wild-type φ6 clones had nonsynonymous mutations in p12, and 2 others had point mutations in p9 and p5. None of these genes had previously been associated with host range expansion in φ6. We demonstrate, for the first time, epistatic constraint in an RNA virus due to host range mutations themselves, which has implications for models of serial host range expansion.IMPORTANCE RNA viruses mutate rapidly and frequently expand their host ranges to infect novel hosts, leading to serial host shifts. Using an RNA bacteriophage model system (Pseudomonas phage φ6), we studied the impact of preexisting host range mutations on another host range expansion. Results from both clonal Sanger and Illumina sequencing show that extant host range mutations dramatically narrow the neighborhood of potential host range mutations compared to that of wild-type φ6. This research suggests that serial host-shifting viruses may follow a small number of molecular paths to enter additional novel hosts. We also identified new genes involved in φ6 host range expansion, expanding our knowledge of this important model system in experimental evolution.

Dual Role of a Viral Polymerase in Viral Genome Replication and Particle Self-Assembly.

Double-stranded RNA (dsRNA) viruses package several RNA-dependent RNA polymerases (RdRp) together with their dsRNA genome into an icosahedral protein capsid known as the polymerase complex. This structure is highly conserved among dsRNA viruses but is not found in any other virus group. RdRp subunits typically interact directly with the main capsid proteins, close to the 5-fold symmetric axes, and perform viral genome replication and transcription within the icosahedral protein shell. In this study, we utilized Pseudomonas phage Φ6, a well-established virus self-assembly model, to probe the potential roles of the RdRp in dsRNA virus assembly. We demonstrated that Φ6 RdRp accelerates the polymerase complex self-assembly process and contributes to its conformational stability and integrity. We highlight the role of specific amino acid residues on the surface of the RdRp in its incorporation during the self-assembly reaction. Substitutions of these residues reduce RdRp incorporation into the polymerase complex during the self-assembly reaction. Furthermore, we determined that the overall transcription efficiency of the Φ6 polymerase complex increased when the number of RdRp subunits exceeded the number of genome segments. These results suggest a mechanism for RdRp recruitment in the polymerase complex and highlight its novel role in virion assembly, in addition to the canonical RNA transcription and replication functions.IMPORTANCE Double-stranded RNA viruses infect a wide spectrum of hosts, including animals, plants, fungi, and bacteria. Yet genome replication mechanisms of these viruses are conserved. During the infection cycle, a proteinaceous capsid, the polymerase complex, is formed. An essential component of this capsid is the viral RNA polymerase that replicates and transcribes the enclosed viral genome. The polymerase complex structure is well characterized for many double-stranded RNA viruses. However, much less is known about the hierarchical molecular interactions that take place in building up such complexes. Using the bacteriophage Φ6 self-assembly system, we obtained novel insights into the processes that mediate polymerase subunit incorporation into the polymerase complex for generation of functional structures. The results presented pave the way for the exploitation and engineering of viral self-assembly processes for biomedical and synthetic biology applications. An understanding of viral assembly processes at the molecular level may also facilitate the development of antivirals that target viral capsid assembly.

Adaptations of an RNA virus to increasing thermal stress.

Environments can change in incremental fashions, where a shift from one state to another occurs over multiple organismal generations. The rate of the environmental change is expected to influence how and how well populations adapt to the final environmental state. We used a model system, the lytic RNA bacteriophage Φ6, to investigate this question empirically. We evolved viruses for thermostability by exposing them to heat shocks that increased to a maximum temperature at different rates. We observed increases in the ability of many heat-shocked populations to survive high temperature heat shocks. On their first exposure to the highest temperature, populations that experienced a gradual increase in temperature had higher average survival than populations that experienced a rapid temperature increase. However, at the end of the experiment, neither the survival of populations at the highest temperature nor the number of mutations per population varied significantly according to the rate of thermal change. We also evaluated mutations from the endpoint populations for their effects on viral thermostability and growth. As expected, some mutations did increase viral thermostability. However, other mutations decreased thermostability but increased growth rate, suggesting that benefits of an increased replication rate may have sometimes outweighed the benefits of enhanced thermostability. Our study highlights the importance of considering the effects of multiple selective pressures, even in environments where a single factor changes.

The emergence of performance trade-offs during local adaptation: insights from experimental evolution.

Environmental heterogeneity is considered a general explanation for phenotypic diversification, particularly when heterogeneity causes populations to diverge via local adaptation. Performance trade-offs, such as those stemming from antagonistic pleiotropy, are thought to contribute to the maintenance of diversity in this scenario. Specifically, alleles that promote adaptation in one environment are expected to promote maladaptation in alternative environments. Contrary to this expectation, however, alleles that underlie locally adaptive traits often fail to exhibit fitness costs in alternative environments. Here, we attempt to explain this paradox by reviewing the results of experimental evolution studies, including a new one of our own, that examined the evolution of trade-offs during adaptation to homogeneous versus heterogeneous environments. We propose that when pleiotropic effects vary, whether or not trade-offs emerge among diverging populations will depend critically on ecology. For example, adaptation to a locally homogeneous environment is more likely to occur by alleles that are antagonistically pleiotropic than adaptation to a locally heterogeneous environment, simply because selection is blind to costs associated with environments that are not experienced locally. Our literature review confirmed the resulting prediction that performance trade-offs were more likely to evolve during selection in homogeneous than heterogeneous environments. The nature of the environmental heterogeneity (spatial versus temporal) and the length of the experiment also contributed in predictable ways to the likelihood that performance trade-offs evolved.

Single-molecule measurements of viral ssRNA packaging.

Genome packaging of double-stranded RNA (dsRNA) phages has been widely studied using biochemical and molecular biology methods. We adapted the existing in vitro packaging system of one such phage for single-molecule experimentation. To our knowledge, this is the first attempt to study the details of viral RNA packaging using optical tweezers. Pseudomonas phage φ6 is a dsRNA virus with a tripartite genome. Positive-sense (+) single-stranded RNA (ssRNA) genome precursors are packaged into a preformed procapsid (PC), where negative strands are synthesized. We present single-molecule measurements of the viral ssRNA packaging by the φ6 PC. Our data show that packaging proceeds intermittently in slow and fast phases, which likely reflects differences in the unfolding of the RNA secondary structures of the ssRNA being packaged. Although the mean packaging velocity was relatively low (0.07-0.54 nm/sec), packaging could reach 4.62 nm/sec during the fast packaging phase.

Oral Immunization of Rabbits with S. enterica Typhimurium Expressing Neisseria gonorrhoeae Filamentous Phage Φ6 Induces Bactericidal Antibodies Against N. gonorrhoeae.

All Neisseria gonorrhoeae strains whose DNA sequences have been determined possess filamentous phage DNA sequences. To ascertain if phage encoded proteins could form the basis of a gonococcal vaccine, rabbits were orally infected with S. enterica Typhimurium strain χ3987 harboring phagemid NgoΦ6 fm. The elicited sera contained large quantities of anti-phage IgG and IgA antibodies that bound to the surface of N. gonorrhoeae cells, as shown by indirect fluorescent analysis and flow cytometry. The elicited sera was able to bind to several phage proteins. The sera also had bactericidal activity. These data demonstrate that N. gonorrhoeae filamentous phage can induce antibodies with anti-gonococcal activity and that phage proteins may be a candidate for vaccine development.

Evolutionary rescue and the coexistence of generalist and specialist competitors: an experimental test.

Competition for resources is thought to play a critical role in both the origins and maintenance of biodiversity. Although numerous laboratory evolution experiments have confirmed that competition can be a key driver of adaptive diversification, few have demonstrated its role in the maintenance of the resulting diversity. We investigate the conditions that favour the origin and maintenance of alternative generalist and specialist resource-use phenotypes within the same population. Previously, we confirmed that competition for hosts among φ6 bacteriophage in a mixed novel (non-permissive) and ancestral (permissive) host microcosm triggered the evolution of a generalist phenotype capable of infecting both hosts. However, because the newly evolved generalists tended to competitively exclude the ancestral specialists, coexistence between the two phenotypes was rare. Here, we show that reducing the relative abundance of the novel host slowed the increase in frequency of the generalist phenotype, allowing sufficient time for the specialist to further adapt to the ancestral host. This adaptation resulted in 'evolutionary rescue' of the specialists, preventing their competitive exclusion by the generalists. Thus, our results suggest that competition promotes both the origin and maintenance of biodiversity when it is strong enough to favour a novel resource-use phenotype, but weak enough to allow adaptation of both the novel and ancestral phenotypes to their respective niches.

Bacteriophage membrane protein P9 as a fusion partner for the efficient expression of membrane proteins in Escherichia coli.

Despite their important roles and economic values, studies of membrane proteins have been hampered by the difficulties associated with obtaining sufficient amounts of protein. Here, we report a novel membrane protein expression system that uses the major envelope protein (P9) of phage φ6 as an N-terminal fusion partner. Phage membrane protein P9 facilitated the synthesis of target proteins and their integration into the Escherichia coli cell membrane. This system was used to produce various multi-pass transmembrane proteins, including G-protein-coupled receptors, transporters, and ion channels of human origin. Green fluorescent protein fusion was used to confirm the correct folding of the expressed proteins. Of the 14 membrane proteins tested, eight were highly expressed, three were moderately expressed, and three were barely expressed in E. coli. Seven of the eight highly expressed proteins could be purified after extraction with the mild detergent lauryldimethylamine-oxide. Although a few proteins have previously been developed as fusion partners to augment membrane protein production, we believe that the major envelope protein P9 described here is better suited to the efficient expression of eukaryotic transmembrane proteins in E. coli.

Frequency and fitness consequences of bacteriophage φ6 host range mutations.

Viruses readily mutate and gain the ability to infect novel hosts, but few data are available regarding the number of possible host range-expanding mutations allowing infection of any given novel host, and the fitness consequences of these mutations on original and novel hosts. To gain insight into the process of host range expansion, we isolated and sequenced 69 independent mutants of the dsRNA bacteriophage Φ6 able to infect the novel host, Pseudomonas pseudoalcaligenes. In total, we found at least 17 unique suites of mutations among these 69 mutants. We assayed fitness for 13 of 17 mutant genotypes on P. pseudoalcaligenes and the standard laboratory host, P. phaseolicola. Mutants exhibited significantly lower fitnesses on P. pseudoalcaligenes compared to P. phaseolicola. Furthermore, 12 of the 13 assayed mutants showed reduced fitness on P. phaseolicola compared to wildtype Φ6, confirming the prevalence of antagonistic pleiotropy during host range expansion. Further experiments revealed that the mechanistic basis of these fitness differences was likely variation in host attachment ability. In addition, using computational protein modeling, we show that host-range expanding mutations occurred in hotspots on the surface of the phage's host attachment protein opposite a putative hydrophobic anchoring domain.

Dominance effects of deleterious and beneficial mutations in a single gene of the RNA virus ϕ6.

Most of our knowledge of dominance stems from studies of deleterious mutations. From these studies we know that most deleterious mutations are recessive, and that this recessivity arises from a hyperbolic relationship between protein function (i.e., protein concentration or activity) and fitness. Here we investigate whether this knowledge can be used to make predictions about the dominance of beneficial and deleterious mutations in a single gene. We employed a model system--the bacteriophage φ6--that allowed us to generate a collection of mutations in haploid conditions so that it was not biased toward either dominant beneficial or recessive deleterious mutations. Screening for the ability to infect a bacterial host that does not permit infection by the wildtype φ6, we generated a collection of mutations in P3, a gene involved in attachment to the host and in phage particle assembly. The resulting collection contained mutations with both deleterious and beneficial effects on fitness. The deleterious mutations in our collection had additive effects on fitness and the beneficial mutations were recessive. Neither of these observations were predicted from previous studies of dominance. This pattern is not consistent with the hyperbolic (diminishing returns) relationship between protein function and fitness that is characteristic of enzymatic genes, but could have resulted from a curve of increasing returns.

Electrostatic interactions drive the self-assembly and the transcription activity of the Pseudomonas phage ϕ6 procapsid.

Assembly of an empty procapsid is a crucial step in the formation of many complex viruses. Here, we used the self-assembly system of the double-stranded RNA bacteriophage ϕ6 to study the role of electrostatic interactions in a scaffolding-independent procapsid assembly pathway. We demonstrate that ϕ6 procapsid assembly is sensitive to salt at both the nucleation and postnucleation steps. Furthermore, we observed that the salt sensitivity of ϕ6 procapsid-directed transcription is reversible.

Rescue of maturation off-pathway products in the assembly of Pseudomonas phage φ 6.

Many complex viruses use an assembly pathway in which their genome is packaged into an empty procapsid which subsequently matures into its final expanded form. We utilized Pseudomonas phage 6, a well-established virus assembly model, to probe the plasticity of the procapsid maturation pathway. The 6 packaging nucleoside triphosphatase (NTPase), which powers sequential translocation of the three viral genomic single-stranded RNA molecules to the procapsid during capsid maturation, is part of the mature 6 virion but may spontaneously be dissociated from the procapsid shell. We demonstrate that the dissociation of NTPase subunits results in premature capsid expansion, which is detected as a change in the sedimentation velocity and as defects in RNA packaging and transcription activity. However, this dead-end conformation of the procapsids was rescued by the addition of purified NTPase hexamers, which efficiently associated on the NTPase-deficient particles and subsequently drove their contraction to the compact naive conformation. The resulting particles regained their biological and enzymatic activities, directing them into a productive maturation pathway. These observations imply that the maturation pathways of complex viruses may contain reversible steps that allow the rescue of the off-pathway conformation in an overall unidirectional virion assembly pathway. Furthermore, we provide direct experimental evidence that particles which have different physical properties (distinct sedimentation velocities and conformations) display different stages of the genome packaging program and show that the transcriptional activity of the 6 procapsids correlates with the number of associated NTPase subunits.

Electrophoretic mobility confirms reassortment bias among geographic isolates of segmented RNA phages.

BACKGROUND: Sex presents evolutionary costs and benefits, leading to the expectation that the amount of genetic exchange should vary in conditions with contrasting cost-benefit equations. Like eukaryotes, viruses also engage in sex, but the rate of genetic exchange is often assumed to be a relatively invariant property of a particular virus. However, the rates of genetic exchange can vary within one type of virus according to geography, as highlighted by phylogeographic studies of cystoviruses. Here we merge environmental microbiology with experimental evolution to examine sex in a diverse set of cystoviruses, consisting of the bacteriophage ϕ6 and its relatives. To quantify reassortment we manipulated - by experimental evolution - electrophoretic mobility of intact virus particles for use as a phenotypic marker to estimate genetic exchange. RESULTS: We generated descendants of ϕ6 that exhibited fast and slow mobility during gel electrophoresis. We identified mutations associated with slow and fast phenotypes using whole genome sequencing and used crosses to establish the production of hybrids of intermediate mobility. We documented natural variation in electrophoretic mobility among environmental isolates of cystoviruses and used crosses against a common fast mobility ϕ6 strain to monitor the production of hybrids with intermediate mobility, thus estimating the amount of genetic exchange. Cystoviruses from different geographic locations have very different reassortment rates when measured against ϕ6, with viruses isolated from California showing higher reassortment rates than those from the Northeastern US. CONCLUSIONS: The results confirm that cystoviruses from different geographic locations have remarkably different reassortment rates -despite similar genome structure and replication mechanisms- and that these differences are in large part due to sexual reproduction. This suggests that particular viruses may indeed exhibit diverse sexual behavior, but wide geographic sampling, across varying environmental conditions may be necessary to characterize the full repertoire. Variation in reassortment rates can assist in the delineation of viral populations and is likely to provide insight into important viral evolutionary dynamics including the rate of coinfection, virulence, and host range shifts. Electrophoretic mobility may be an indicator of important determinants of fitness and the techniques herein can be applied to the study of other viruses.

Migration enhances adaptation in bacteriophage populations evolving in ecological sinks.

Migration between populations can be a major evolutionary force. However, some disagreement exists as to precisely how migration affects population adaptation. Some theories emphasize the inhibitory effects of gene flow between locally adapted populations, whereas others propose that migration can enhance adaptation. Migration has also been theorized to rescue sink populations from extinction. In our experiments, we serially passaged bacteriophage Φ6 host range mutants under sink conditions on a novel host while manipulating the source and number of migrants into these experimental populations. Migrants from two sources were used: mutant Φ6 phage able to infect a novel host (treatment) and wild-type Φ6 phage unable to infect a novel host (control). We used quadratic regressions to determine the relationship between the number of migrants per passage and the absolute fitnesses of experimental populations following 30 passages. Our results showed that migration from a control population had no effect on absolute fitnesses of our serially passaged populations following 30 passages. By contrast, the relationship between migrants per passage and absolute fitnesses for populations receiving migrants able to infect the novel host was best described by an upwardly concave curve. These results suggest that intermediate levels of migration can have favorable impacts on evolutionary adaptation.

Competition and the origins of novelty: experimental evolution of niche-width expansion in a virus.

Competition for resources has long been viewed as a key agent of divergent selection. Theory holds that populations facing severe intraspecific competition will tend to use a wider range of resources, possibly even using entirely novel resources that are less in demand. Yet, there have been few experimental tests of these ideas. Using the bacterial virus (bacteriophage) 6 as a model system, we examined whether competition for host resources promotes the evolution of novel resource use. In the laboratory, 6 exhibits a narrow host range but readily produces mutants capable of infecting novel bacterial hosts. Here, we show that when 6 populations were subjected to intense intraspecific competition for their standard laboratory host, they rapidly evolved new generalist morphs that infect novel hosts. Our results therefore suggest that competition for host resources may drive the evolution of host range expansion in viruses. More generally, our findings demonstrate that intraspecific resource competition can indeed promote the evolution of novel resource-use phenotypes.

Protein P7 of the cystovirus φ6 is located at the three-fold axis of the unexpanded procapsid.

The objective of this study was to determine the location of protein P7, the RNA packaging factor, in the procapsid of the φ6 cystovirus. A comparison of cryo-electron microscopy high-resolution single particle reconstructions of the φ6 complete unexpanded procapsid, the protein P2-minus procapsid (P2 is the RNA directed RNA-polymerase), and the P7-minus procapsid, show that prior to RNA packaging the P7 protein is located near the three-fold axis of symmetry. Difference maps highlight the precise position of P7 and demonstrate that in P7-minus particles the P2 proteins are less localized with reduced densities at the three-fold axes. We propose that P7 performs the mechanical function of stabilizing P2 on the inner protein P1 shell which ensures that entering viral single-stranded RNA is replicated.

Evolutionary genomics of host-use in bifurcating demes of RNA virus phi-6.

BACKGROUND: Viruses are exceedingly diverse in their evolved strategies to manipulate hosts for viral replication. However, despite these differences, most virus populations will occasionally experience two commonly-encountered challenges: growth in variable host environments, and growth under fluctuating population sizes. We used the segmented RNA bacteriophage ϕ6 as a model for studying the evolutionary genomics of virus adaptation in the face of host switches and parametrically varying population sizes. To do so, we created a bifurcating deme structure that reflected lineage splitting in natural populations, allowing us to test whether phylogenetic algorithms could accurately resolve this 'known phylogeny'. The resulting tree yielded 32 clones at the tips and internal nodes; these strains were fully sequenced and measured for phenotypic changes in selected traits (fitness on original and novel hosts). RESULTS: We observed that RNA segment size was negatively correlated with the extent of molecular change in the imposed treatments; molecular substitutions tended to cluster on the Small and Medium RNA chromosomes of the virus, and not on the Large segment. Our study yielded a very large molecular and phenotypic dataset, fostering possible inferences on genotype-phenotype associations. Using further experimental evolution, we confirmed an inference on the unanticipated role of an allelic switch in a viral assembly protein, which governed viral performance across host environments. CONCLUSIONS: Our study demonstrated that varying complexities can be simultaneously incorporated into experimental evolution, to examine the combined effects of population size, and adaptation in novel environments. The imposed bifurcating structure revealed that some methods for phylogenetic reconstruction failed to resolve the true phylogeny, owing to a paucity of molecular substitutions separating the RNA viruses that evolved in our study.