Enduring questions in regenerative biology and the search for answers.

The potential for basic research to uncover the inner workings of regenerative processes and produce meaningful medical therapies has inspired scientists, clinicians, and patients for hundreds of years. Decades of studies using a handful of highly regenerative model organisms have significantly advanced our knowledge of key cell types and molecular pathways involved in regeneration. However, many questions remain about how regenerative processes unfold in regeneration-competent species, how they are curtailed in non-regenerative organisms, and how they might be induced (or restored) in humans. Recent technological advances in genomics, molecular biology, computer science, bioengineering, and stem cell research hold promise to collectively provide new experimental evidence for how different organisms accomplish the process of regeneration. In theory, this new evidence should inform the design of new clinical approaches for regenerative medicine. A deeper understanding of how tissues and organs regenerate will also undoubtedly impact many adjacent scientific fields. To best apply and adapt these new technologies in ways that break long-standing barriers and answer critical questions about regeneration, we must combine the deep knowledge of developmental and evolutionary biologists with the hard-earned expertise of scientists in mechanistic and technical fields. To this end, this perspective is based on conversations from a workshop we organized at the Banbury Center, during which a diverse cross-section of the regeneration research community and experts in various technologies discussed enduring questions in regenerative biology. Here, we share the questions this group identified as significant and unanswered, i.e., known unknowns. We also describe the obstacles limiting our progress in answering these questions and how expanding the number and diversity of organisms used in regeneration research is essential for deepening our understanding of regenerative capacity. Finally, we propose that investigating these problems collaboratively across a diverse network of researchers has the potential to advance our field and produce unexpected insights into important questions in related areas of biology and medicine.

RNAi Screen of RING/U-Box Domain Ubiquitin Ligases Identifies Critical Regulators of Tissue Regeneration in Planarians.

Regenerative processes depend on the interpretation of signals to coordinate cell behaviors. The role of ubiquitin-mediated signaling is known to be important in many cellular and biological contexts, but its role in regeneration is not well understood. To investigate how ubiquitylation impacts tissue regeneration in vivo, we are studying planarians that are capable of regenerating after nearly any injury using a population of stem cells. Here we used RNAi to screen RING/U-box E3 ubiquitin ligases that are highly expressed in planarian stem cells and stem cell progeny. RNAi screening identified nine genes with functions in regeneration, including the spliceosomal factor prpf19 and histone modifier rnf2; based on their known roles in developmental processes, we further investigated these two genes. We found that prpf19 was required for animal survival but not for stem cell maintenance, suggesting a role in promoting cell differentiation. Because RNF2 is the catalytic subunit of the Polycomb Repressive Complex 1 (PRC1), we also examined other putative members of this complex (CBX and PHC). We observed a striking phenotype of regional tissue misspecification in cbx and phc RNAi planarians. To identify genes regulated by PRC1, we performed RNA-seq after knocking down rnf2 or phc. Although these proteins are predicted to function in the same complex, we found that the set of genes differentially expressed in rnf2 versus phc RNAi were largely non-overlapping. Using in situ hybridization, we showed that rnf2 regulates gene expression levels within a tissue type, whereas phc is necessary for the spatial restriction of gene expression, findings consistent with their respective in vivo phenotypes. This work not only uncovered roles for RING/U-box E3 ligases in stem cell regulation and regeneration, but also identified differential gene targets for two putative PRC1 factors required for maintaining cell-type-specific gene expression in planarians.

A modular laboratory course using planarians to study genes involved in tissue regeneration.

Undergraduate research experiences are excellent opportunities to engage students in science alongside experienced scientists, but at large institutions, it is challenging to accommodate all students. To address and engage a larger number of students, we developed a modular laboratory course based on the course-based undergraduate research experiences model. This new course was integrated with the scientific aims of a research laboratory studying the cellular and molecular mechanisms underlying tissue regeneration in planarians. In this course, students were asked to identify genes with roles in planarian biology. Students analyzed and cloned an assigned gene, determined its expression pattern in situ and examined its function in regeneration. Additionally, we developed critical thinking and scientific communication skills by incorporating activities focused on critical concepts. Students obtained high quality primary data and were successful in completing and mastering the course learning outcomes. They benefitted by developing basic research skills, learning to perform, trouble-shooting experiments, reading and critically analyzing primary literature, and using the information to defend and explain their experimental results. Through this course, students also increased their confidence and ability to perform independent scientific research. The course was designed to make it accessible to the community to implement and adapt as appropriate in diverse institutions. © 2019 International Union of Biochemistry and Molecular Biology, 47(5):547-559, 2019.

Post-translational regulation of planarian regeneration.

Most mammals cannot easily overcome degenerative disease or traumatic injuries. In contrast, an innate ability to regenerate is observed across animal phyla. Freshwater planarians are amongst the organisms that are capable of stem cell-mediated whole-body regeneration and have served as an exemplary model to study how pluripotency is maintained and regulated in vivo. Here, we review findings on the role of post-translational modifications and the genes regulating phosphorylation, ubiquitylation, and chromatin remodeling in planarian regeneration. Furthermore, we discuss how technological advances for identifying cellular targets of these processes will fill gaps in our knowledge of the signaling mechanisms that underlie regeneration in planarians, which should inform how tissue repair can be stimulated in non-regenerative model organisms and in humans.

Planarians as models to investigate the bioactivity of gold(I) complexes in vivo.

Gold(I)-containing complexes are used in drug discovery research for rheumatoid arthritis, cancer, and parasitic infections. In this study, we tested the bioactivity of gold(I) complexes in vivo using planarians. The planarian Schmidtea mediterranea possesses orthologues of tumor suppressor genes, such as p53, that, when silenced, cause deregulation of cell proliferation and apoptosis. In this context, we tested two triethylphosphine-gold(I) complexes (AdO and AdT) to determine if they can attenuate phenotypes that result from p53 inhibition. First, we identified the drug concentration that did not affect survival or regeneration and evaluated the drug's effect on cell division and apoptosis. We found that AdT treatment decreased the number of mitotic cells and that all drug treatments increased the number of apoptotic cells. We then performed p53(RNAi) and drug treatments concomitantly and observed the phenotype progression. Drug treatment increased survival three-fold and decreased apoptosis, which resulted in an attenuated phenotype. Our results indicate that planarians can be treated with gold(I) complexes, and that this treatment can diminish the p53(RNAi) phenotype and extend survival. In this work we show that planarians can be used as a model to study the in vivo effect of gold(I) complexes and to further investigate their mechanisms of action.

SoxB1 Activity Regulates Sensory Neuron Regeneration, Maintenance, and Function in Planarians.

SoxB1 genes play fundamental roles in neurodevelopmental processes and maintaining stem cell multipotency, but little is known about their function in regeneration. We addressed this question by analyzing the activity of the SoxB1 homolog soxB1-2 in the planarian Schmidtea mediterranea. Expression and functional analysis revealed that soxB1-2 marks ectodermal-lineage progenitors, and its activity is required for differentiation of subsets of ciliated epidermal and neuronal cells. Moreover, we show that inhibiting soxB1-2 or its candidate target genes leads to abnormal sensory neuron regeneration that causes planarians to display seizure-like movements or phenotypes associated with the loss of sensory modalities. Our analyses highlight soxB1-2-regulated genes that are expressed in sensory neurons and are homologous to factors implicated in epileptic disorders in humans and animal models of epilepsy, indicating that planarians can serve as a complementary model to investigate genetic causes of epilepsy.

Fixation, Processing, and Immunofluorescent Labeling of Whole Mount Planarians.

Efforts to elucidate mechanisms of regeneration in the planarian Schmidtea mediterranea have included the application of immunocytochemical methods to detect specific molecules and label cells and tissues in situ. Here we describe methods for immunofluorescent labeling of whole mount planarians. We outline protocols for fixation and steps for processing animals prior to immunolabeling, incorporating commonly utilized reagents for mucus removal, pigment bleaching, tissue permeabilization, and antigen retrieval. Because processing steps can mask or degrade antigens, we also recommend protocol parameters that can be tested simultaneously to optimize sample preparation for novel antibodies.

Dissecting the function of Cullin-RING ubiquitin ligase complex genes in planarian regeneration.

The ubiquitin system plays a role in nearly every aspect of eukaryotic cell biology. The enzymes responsible for transferring ubiquitin onto specific substrates are the E3 ubiquitin ligases, a large and diverse family of proteins, for which biological roles and target substrates remain largely undefined. Studies using model organisms indicate that ubiquitin signaling mediates key steps in developmental processes and tissue regeneration. Here, we used the freshwater planarian, Schmidtea mediterranea, to investigate the role of Cullin-RING ubiquitin ligase (CRL) complexes in stem cell regulation during regeneration. We identified six S. mediterranea cullin genes, and used RNAi to uncover roles for homologs of Cullin-1, -3 and -4 in planarian regeneration. The cullin-1 RNAi phenotype included defects in blastema formation, organ regeneration, lesions, and lysis. To further investigate the function of cullin-1-mediated cellular processes in planarians, we examined genes encoding the adaptor protein Skp1 and F-box substrate-recognition proteins that are predicted to partner with Cullin-1. RNAi against skp1 resulted in phenotypes similar to cullin-1 RNAi, and an RNAi screen of the F-box genes identified 19 genes that recapitulated aspects of cullin-1 RNAi, including ones that in mammals are involved in stem cell regulation and cancer biology. Our data provides evidence that CRLs play discrete roles in regenerative processes and provide a platform to investigate how CRLs regulate stem cells in vivo.

Nervous system development and regeneration in freshwater planarians.

Planarians have a long history in the fields of developmental and regenerative biology. These animals have also sparked interest in neuroscience due to their neuroanatomy, spectrum of simple behaviors, and especially, their almost unparalleled ability to generate new neurons after any type of injury. Research in adult planarians has revealed that neuronal subtypes homologous to those found in vertebrates are generated from stem cells throughout their lives. This feat is recapitulated after head amputation, wherein animals are capable of regenerating whole brains and regaining complete neural function. In this review, we summarize early studies on the anatomy and function of the planarian nervous system and discuss our present knowledge of the molecular mechanisms governing neurogenesis in planarians. Modern studies demonstrate that the transcriptional programs underlying neuronal specification are conserved in these remarkable organisms. Thus, planarians are outstanding models to investigate questions about how stem cells can replace neurons in vivo. WIREs Dev Biol 2017, 6:e266. doi: 10.1002/wdev.266 For further resources related to this article, please visit the WIREs website.

A functional genomics screen identifies an Importin-α homolog as a regulator of stem cell function and tissue patterning during planarian regeneration.

BACKGROUND: Planarians are renowned for their regenerative capacity and are an attractive model for the study of adult stem cells and tissue regeneration. In an effort to better understand the molecular mechanisms underlying planarian regeneration, we performed a functional genomics screen aimed at identifying genes involved in this process in Schmidtea mediterranea. METHODS: We used microarrays to detect changes in gene expression in regenerating and non-regenerating tissues in planarians regenerating one side of the head and followed this with high-throughput screening by in situ hybridization and RNAi to characterize the expression patterns and function of the differentially expressed genes. RESULTS: Along with five previously characterized genes (Smed-cycD, Smed-morf41/mrg-1, Smed-pdss2/dlp1, Smed-slbp, and Smed-tph), we identified 20 additional genes necessary for stem cell maintenance (Smed-sart3, Smed-smarcc-1, Smed-espl1, Smed-rrm2b-1, Smed-rrm2b-2, Smed-dkc1, Smed-emg1, Smed-lig1, Smed-prim2, Smed-mcm7, and a novel sequence) or general regenerative capability (Smed-rbap46/48-2, Smed-mcm2, Smed-ptbp1, and Smed-fen-1) or that caused tissue-specific defects upon knockdown (Smed-ddc, Smed-gas8, Smed-pgbd4, and Smed-b9d2). We also found that a homolog of the nuclear transport factor Importin-α plays a role in stem cell function and tissue patterning, suggesting that controlled nuclear import of proteins is important for regeneration. CONCLUSIONS: Through this work, we described the roles of several previously uncharacterized genes in planarian regeneration and implicated nuclear import in this process. We have additionally created an online database to house our in situ and RNAi data to make it accessible to the planarian research community.

Identification of HECT E3 ubiquitin ligase family genes involved in stem cell regulation and regeneration in planarians.

E3 ubiquitin ligases constitute a large family of enzymes that modify specific proteins by covalently attaching ubiquitin polypeptides. This post-translational modification can serve to regulate protein function or longevity. In spite of their importance in cell physiology, the biological roles of most ubiquitin ligases remain poorly understood. Here, we analyzed the function of the HECT domain family of E3 ubiquitin ligases in stem cell biology and tissue regeneration in planarians. Using bioinformatic searches, we identified 17 HECT E3 genes that are expressed in the Schmidtea mediterranea genome. Whole-mount in situ hybridization experiments showed that HECT genes were expressed in diverse tissues and most were expressed in the stem cell population (neoblasts) or in their progeny. To investigate the function of all HECT E3 ligases, we inhibited their expression using RNA interference (RNAi) and determined that orthologs of huwe1, wwp1, and trip12 had roles in tissue regeneration. We show that huwe1 RNAi knockdown led to a significant expansion of the neoblast population and death by lysis. Further, our experiments showed that wwp1 was necessary for both neoblast and intestinal tissue homeostasis as well as uncovered an unexpected role of trip12 in posterior tissue specification. Taken together, our data provide insights into the roles of HECT E3 ligases in tissue regeneration and demonstrate that planarians will be a useful model to evaluate the functions of E3 ubiquitin ligases in stem cell regulation.

Novel monoclonal antibodies to study tissue regeneration in planarians.

BACKGROUND: Planarians are an attractive model organism for studying stem cell-based regeneration due to their ability to replace all of their tissues from a population of adult stem cells. The molecular toolkit for planarian studies currently includes the ability to study gene function using RNA interference (RNAi) and observe gene expression via in situ hybridizations. However, there are few antibodies available to visualize protein expression, which would greatly enhance analysis of RNAi experiments as well as allow further characterization of planarian cell populations using immunocytochemistry and other immunological techniques. Thus, additional, easy-to-use, and widely available monoclonal antibodies would be advantageous to study regeneration in planarians. RESULTS: We have created seven monoclonal antibodies by inoculating mice with formaldehyde-fixed cells isolated from dissociated 3-day regeneration blastemas. These monoclonal antibodies can be used to label muscle fibers, axonal projections in the central and peripheral nervous systems, two populations of intestinal cells, ciliated cells, a subset of neoblast progeny, and discrete cells within the central nervous system as well as the regeneration blastema. We have tested these antibodies using eight variations of a formaldehyde-based fixation protocol and determined reliable protocols for immunolabeling whole planarians with each antibody. We found that labeling efficiency for each antibody varies greatly depending on the addition or removal of tissue processing steps that are used for in situ hybridization or immunolabeling techniques. Our experiments show that a subset of the antibodies can be used alongside markers commonly used in planarian research, including anti-SYNAPSIN and anti-SMEDWI, or following whole-mount in situ hybridization experiments. CONCLUSIONS: The monoclonal antibodies described in this paper will be a valuable resource for planarian research. These antibodies have the potential to be used to better understand planarian biology and to characterize phenotypes following RNAi experiments. In addition, we present alterations to fixation protocols and demonstrate how these changes can increase the labeling efficiencies of antibodies used to stain whole planarians.

COE loss-of-function analysis reveals a genetic program underlying maintenance and regeneration of the nervous system in planarians.

Members of the COE family of transcription factors are required for central nervous system (CNS) development. However, the function of COE in the post-embryonic CNS remains largely unknown. An excellent model for investigating gene function in the adult CNS is the freshwater planarian. This animal is capable of regenerating neurons from an adult pluripotent stem cell population and regaining normal function. We previously showed that planarian coe is expressed in differentiating and mature neurons and that its function is required for proper CNS regeneration. Here, we show that coe is essential to maintain nervous system architecture and patterning in intact (uninjured) planarians. We took advantage of the robust phenotype in intact animals to investigate the genetic programs coe regulates in the CNS. We compared the transcriptional profiles of control and coe RNAi planarians using RNA sequencing and identified approximately 900 differentially expressed genes in coe knockdown animals, including 397 downregulated genes that were enriched for nervous system functional annotations. Next, we validated a subset of the downregulated transcripts by analyzing their expression in coe-deficient planarians and testing if the mRNAs could be detected in coe+ cells. These experiments revealed novel candidate targets of coe in the CNS such as ion channel, neuropeptide, and neurotransmitter genes. Finally, to determine if loss of any of the validated transcripts underscores the coe knockdown phenotype, we knocked down their expression by RNAi and uncovered a set of coe-regulated genes implicated in CNS regeneration and patterning, including orthologs of sodium channel alpha-subunit and pou4. Our study broadens the knowledge of gene expression programs regulated by COE that are required for maintenance of neural subtypes and nervous system architecture in adult animals.

Genome-wide analysis of the bHLH gene family in planarians identifies factors required for adult neurogenesis and neuronal regeneration.

In contrast to most well-studied model organisms, planarians have a remarkable ability to completely regenerate a functional nervous system from a pluripotent stem cell population. Thus, planarians provide a powerful model to identify genes required for adult neurogenesis in vivo. We analyzed the basic helix-loop-helix (bHLH) family of transcription factors, many of which are crucial for nervous system development and have been implicated in human diseases. However, their potential roles in adult neurogenesis or central nervous system (CNS) function are not well understood. We identified 44 planarian bHLH homologs, determined their patterns of expression in the animal and assessed their functions using RNAi. We found nine bHLHs expressed in stem cells and neurons that are required for CNS regeneration. Our analyses revealed that homologs of coe, hes (hesl-3) and sim label progenitors in intact planarians, and following amputation we observed an enrichment of coe(+) and sim(+) progenitors near the wound site. RNAi knockdown of coe, hesl-3 or sim led to defects in CNS regeneration, including failure of the cephalic ganglia to properly pattern and a loss of expression of distinct neuronal subtype markers. Together, these data indicate that coe, hesl-3 and sim label neural progenitor cells, which serve to generate new neurons in uninjured or regenerating animals. Our study demonstrates that this model will be useful to investigate how stem cells interpret and respond to genetic and environmental cues in the CNS and to examine the role of bHLH transcription factors in adult tissue regeneration.

Epigenetic regulation of planarian stem cells by the SET1/MLL family of histone methyltransferases.

Chromatin regulation is a fundamental mechanism underlying stem cell pluripotency, differentiation, and the establishment of cell type-specific gene expression profiles. To examine the role of chromatin regulation in stem cells in vivo, we study regeneration in the freshwater planarian Schmidtea mediterranea. These animals possess a high concentration of pluripotent stem cells, which are capable of restoring any damaged or lost tissues after injury or amputation. Here, we identify the S. mediterranea homologs of the SET1/MLL family of histone methyltransferases and COMPASS and COMPASS-like complex proteins and investigate their role in stem cell function during regeneration. We identified six S. mediterranea homologs of the SET1/MLL family (set1, mll1/2, trr-1, trr-2, mll5-1 and mll5-2), characterized their patterns of expression in the animal, and examined their function by RNAi. All members of this family are expressed in the stem cell population and differentiated tissues. We show that set1, mll1/2, trr-1, and mll5-2 are required for regeneration and that set1, trr-1 and mll5-2 play roles in the regulation of mitosis. Most notably, knockdown of the planarian set1 homolog leads to stem cell depletion. A subset of planarian homologs of COMPASS and COMPASS-like complex proteins are also expressed in stem cells and implicated in regeneration, but the knockdown phenotypes suggest that some complex members also function in other aspects of planarian biology. This work characterizes the function of the SET1/MLL family in the context of planarian regeneration and provides insight into the role of these enzymes in adult stem cell regulation in vivo.

A Lissencephaly-1 homologue is essential for mitotic progression in the planarian Schmidtea mediterranea.

BACKGROUND: Planarians are renowned for their capacity to replace lost tissues from adult pluripotent stem cells (neoblasts). Here we report that Lissencephaly-1 (lis1), which has roles in cellular processes such as mitotic spindle apparatus orientation and in signal regulation required for stem cell self-renewal, is required for stem cell maintenance in the planarian Schmidtea mediterranea. RESULTS: In planarians, lis1 is expressed in differentiated tissues and stem cells. lis1 RNAi leads to head regression, ventral curling, and death by lysis. By labeling the neoblasts and proliferating cells, we found lis1 knockdown animals show a dramatic increase in the number of mitotic cells, followed by depletion of the stem cell pool. Analysis of the mitotic spindles in dividing neoblasts revealed that defective spindle positioning is correlated with cells arrested at metaphase. In addition, we show that inhibiting a planarian homologue of nudE, predicted to encode a LIS-1 interacting protein, also leads to cell cycle progression defects. CONCLUSIONS: Our results provide evidence for a conserved role of LIS1 and NUDE in regulating the function of the mitotic spindle apparatus in a representative Lophotrochozoan and that planarians will be useful organisms in which to investigate LIS1 regulation of signaling events underlying stem cell self-renewal.

The use of lectins as markers for differentiated secretory cells in planarians.

Freshwater planarians have reemerged as excellent models to investigate mechanisms underlying regeneration. The introduction of molecular tools has facilitated the study of planarians, but cell- and tissue-specific markers are still needed to examine differentiation of most cell types. Here we report the utility of fluorescent lectin-conjugates to label tissues in the planarian Schmidtea mediterranea. We show that 16 lectin-conjugates stain planarian cells or tissues; 13 primarily label the secretory cells, their cytoplasmic projections, and terminal pores. Thus, we examined regeneration of the secretory system using lectin markers and functionally characterized two genes expressed in the secretory cells: marginal adhesive gland-1 (mag-1) and Smed-reticulocalbin1 (Smed-rcn1). RNAi knockdown of these genes caused a dramatic reduction of secretory cell lectin staining, suggesting a role for mag-1 and Smed-rcn1 in secretory cell differentiation. Our results provide new insights into planarian secretory system regeneration and add new markers for labeling several planarian tissues.

nanos function is essential for development and regeneration of planarian germ cells.

Germ cells are required for the successful propagation of sexually reproducing species. Understanding the mechanisms by which these cells are specified and how their totipotency is established and maintained has important biomedical and evolutionary implications. Freshwater planarians serve as fascinating models for studying these questions. They can regenerate germ cells from fragments of adult tissues that lack reproductive structures, suggesting that inductive signaling is involved in planarian germ cell specification. To study the development and regeneration of planarian germ cells, we have functionally characterized an ortholog of nanos, a gene required for germ cell development in diverse organisms, from Schmidtea mediterranea. In the hermaphroditic strain of this species, Smed-nanos mRNA is detected in developing, regenerating, and mature ovaries and testes. However, it is not detected in the vast majority of newly hatched planarians or in small tissue fragments that will ultimately regenerate germ cells, consistent with an epigenetic origin of germ cells. We show that Smed-nanos RNA interference (RNAi) results in failure to develop, regenerate, or maintain gonads in sexual planarians. Unexpectedly, Smed-nanos mRNA is also detected in presumptive testes primordia of asexual individuals that reproduce strictly by fission. These presumptive germ cells are lost after Smed-nanos RNAi, suggesting that asexual planarians specify germ cells, but their differentiation is blocked downstream of Smed-nanos function. Our results reveal a conserved function of nanos in germ cell development in planarians and suggest that these animals will serve as useful models for dissecting the molecular basis of epigenetic germ cell specification.

Characterization of NO/cGMP-mediated responses in identified motoneurons.

1. Nitric oxide (NO) is thought to play a neuromodulatory role in the nervous system of vertebrate and invertebrate species. In the hornworm Manduca sexta, NO-mediated signaling has been implicated in behavioral and developmental processes, but its exact function in neurons is unknown. In this study, we identify specific neurons in the CNS of Manduca larvae that accumulate cGMP in response to treatment with NO donors in the presence of cGMP-phosphodiesterase inhibitors. Subsets of these neurons were identified as motoneuron-12 (MN12) and intersegmental motoneurons (ISMs), which innervate dorsal oblique muscles of the larvae. 2. To investigate the physiological role of NO-evoked increases in cGMP in these motoneurons we performed intracellular recordings; we found that application of NO donors caused an increase in neuronal excitability that was characterized by an increase in the spontaneous firing frequency. When action potentials and EPSPs were blocked, NO treatment evoked a depolarization of the resting membrane potential and a decrease in the measured input resistance in both MN12 and the ISMs. 3. Additional experiments with MN12 showed that treatment with the cGMP analogue, 8-Br-cGMP mimicked the NO effect on the resting potential and the input resistance. Furthermore, MN12 incubation with the NOS inhibitor, L-NNA, resulted in a small hyperpolarization of the resting potential and an increase in the input resistance, and incubation with the sGC inhibitor, ODQ blocked the NO-evoked depolarization of MN12. Finally, NO treatment during voltage clamping of MN12 evoked an inward positive current. 4. Taken together, these results suggest that NO can act as a "gain control" of neuronal excitability, which might have an important role in insect behavior.