We sought genes involved in processes that regulate where fate-specified progenitors differentiate through RNAi studies. We found a gene (map3k1, dd_5198) encoding a MAP3K1-like signaling protein for which RNAi resulted in a striking phenotype involving dispersed ectopic eyes and single eye cells (Figure 1A). The unique nature of this patterning phenotype raised the possibility that map3k1 has some role in regulating progenitor differentiation during targeting. Planarian map3k1 encodes a member of the MAP3K signaling protein family. Aside from a characteristic kinase domain, eukaryotic MAP3K1 proteins, including the planarian ortholog, possess unique accessory domains compared to other MAP3Ks; these include a PHD-like RING finger, a SWIM-type RING finger, and a TOG-like domain. These domains have been implicated in non-canonical MAP kinase signaling cascades that enable MAP3K1 to act as a ubiquitin ligase and a scaffold protein (Lu et al., 2002; Xia et al., 2007; Figure 1—figure supplement 1A).

The bilaterally symmetric anterior location of planarian eyes represents the normal targeting location of eye progenitors (Lapan and Reddien, 2011; Lapan and Reddien, 2012). RNAi of map3k1 resulted in the gradual emergence of ectopic eyes posterior to the normal eye location with variability in their anterior–posterior (AP) and medial–lateral (ML) positions (Figure 1A). A similar map3k1 RNAi phenotype was independently described in Lo and Petersen, 2025. Fluorescent in situ hybridization (FISH) experiments showed that both optic cup (OC) cells and photoreceptor neurons (PRNs) were present ectopically in clusters of cells and as individual, isolated cells (Figure 1A). Isolated, individual PRN or OC cell differentiation is rarely observed in the wild-type state. Although ectopic eye cells appeared most frequently in the normal zone of eye-progenitor specification (the head), they also appeared throughout the trunk and tail – far from the canonical eye-progenitor specification zone (Figure 1B, C; Figure 1—figure supplement 1B). Isolated eye cells were observed in animal tails within 1–2 weeks of the first RNAi feeding when no, or very few other ectopic eye cells appeared between the head and tail (Figure 1C, D; Figure 1—figure supplement 1B). The far posterior ectopic eye cells that appeared in map3k1 RNAi animals were sparser than those in the head and usually appeared as singletons rather than aggregates (Figure 1C; Figure 1—figure supplement 1B).

The precise location in the planarian body where dispersed eye progenitors migrate toward and target for differentiation is referred to as the target zone (TZ) (Atabay et al., 2018). A variety of genes can be inhibited to result in eye formation in ectopic positions along the AP and ML axes and are presumed to impact the location of the TZ: for instance, posterior TZ shifting following RNAi of nou darake (ndk) (Cebrià et al., 2002) and wntA (Kobayashi et al., 2007); anterior shifting following RNAi of notum (Hill and Petersen, 2015) and nr4A (Li et al., 2019); lateral shifting following RNAi of wnt5; and medial shifting following RNAi of slit (Oderberg et al., 2017; Atabay et al., 2018). However, in all of these cases, organized ectopic eyes appear along particular AP or ML trajectories (Figure 1E, summary cartoon), consistent with alteration of the eye-progenitor TZ along a single axis, but still with differentiation constrained to occur in a particular position on the orthogonal axis. By contrast, map3k1 RNAi ectopic eyes and eye cells were more disorganized, often being ectopic in both AP and ML position (Figure 1A). This distinct phenotype raised the possibility that map3k1 is required for some patterning process not previously disrupted by the inhibition of other patterning genes.

Patterning in planarians prominently involves constitutive and regional expression of genes constituting positional information, referred to as position control genes (PCGs) (Reddien, 2018). PCGs are predominantly expressed in planarian muscle (Witchley et al., 2013). ndk, notum, nr4A, wnt5, and slit are regulated in their spatial expression and are components of this PCG patterning system. Therefore, these genes likely influence progenitor-extrinsic cues that guide progenitors to particular locations. By contrast, map3k1 was not overtly expressed in a spatially restricted manner; instead, it was expressed broadly across tissues (Figure 1—figure supplement 1C).

The appearance of eye cells in the tail of map3k1 RNAi animals suggests that some eye-progenitor specification likely occurred in the tails of these animals. Eye-specialized neoblasts can be recognized by the co-expression of the eye-specific FSTF ovo (Lapan and Reddien, 2012) and the neoblast marker smedwi-1 (Reddien et al., 2005). ovo⁺; smedwi-1⁺ neoblasts were observed in the midbody and tail of map3k1 RNAi animals (Figure 1F; Figure 1—figure supplement 1D, E). However, the frequency of observed ovo⁺ neoblasts in the tail was very low, and a similar low frequency of ovo⁺ neoblasts was observed in control animal tails (Figure 1F; Figure 1—figure supplement 1E). There was therefore no overt change to the spatial pattern of eye-specialized neoblasts after map3k1 RNAi. This raises the possibility that there exists a natural low frequency of sporadic eye-neoblast specification events outside of the predominant, anterior eye-progenitor specification zone and that these cells can ectopically differentiate into mature eye cells in the posterior when map3k1 is inhibited.

To assess whether the ectopic pattern of differentiated cells in map3k1 RNAi animals was specific to the eye, we visualized other differentiated tissue patterns. Laterally projecting neural branches in the central nervous system are normally restricted to the anterior planarian brain (Hyman, 1951). After map3k1 RNAi, however, ectopic brain branches labeled by the neuronal markers choline acetyltransferase (chat) and prohormone convertase-2 (pc2) emerged from the two ventral nerve cords, appearing most frequently in the anterior and mid-body of the animal (Figure 2A; Figure 2—figure supplement 1A). In some instances, branches were observed in the tail (Figure 2—figure supplement 1B). Ectopic eyes sent axons (labeled with an anti-Arrestin antibody) that traveled along these ectopic neural branches and main nerve cord tracts (Figure 2A; Figure 2—figure supplement 1A, B). Ectopic brain branches that emerged from ventral nerve cords contained GluR⁺ (dd_16476) neurons, confirming that they are at least partly composed of normally brain-branch-restricted neurons (Figure 2B). GluR⁺ neurons are similarly found in ectopic brain branches observed after RNAi of ndk (Cebrià et al., 2002).

The anterior-restricted population of dd_17258⁺ neurons also displayed ectopic differentiation across the AP axis of map3k1 RNAi animals, extending to the tail tip (Figure 2C; Figure 2—figure supplement 1C, D). The correct number of dd_17258⁺ neurons, however, remained in the normal AP location (AP_1) (Figure 2D). By contrast, some other anterior neural populations, cintillo⁺ and glutamic acid decarboxylase⁺ (gad⁺) neurons, remained unaffected in map3k1 RNAi animals (Figure 2C; Figure 2—figure supplement 1C). These findings suggest that map3k1 is required for the normal AP restriction of a subset of neural cell types during tissue turnover.

To determine if tissue posteriorization in map3k1 RNAi occurred for non-neuronal cell types other than OC cells, we assayed gland cell populations (dd_9223, dd_7131⁺, and dd_8476⁺) that normally reside in the head, with a small fraction extending posteriorly to the pharynx and rarely to the tail. map3k1 RNAi animals showed posteriorization of gland cell distributions for the primarily anterior dd_7131⁺and dd_8476⁺ populations (Figure 2E). As was the case with neuron types, not all gland cell types were affected by map3k1 RNAi. dd_9223⁺ cells, the most anteriorly restricted of the three gland cell populations assessed, did not change in distribution (Figure 2E). These data indicate that map3k1 broadly affects the spatial distribution of various differentiated cells during tissue maintenance, but that this role is restricted to a subset of cell types.

Another regional tissue that is maintained through turnover from regional progenitors is the planarian pharynx. Neoblasts that produce pharynx progenitors are broadly located in the trunk region of the animal and express FoxA (Adler et al., 2014; Scimone et al., 2014a). Pharynx progenitors enter the pharynx through a connection to the body at the anterior end of the organ, and this requires that progenitors are capable of moving in multiple directions as a response to extrinsic cues. After 3 weeks of map3k1 RNAi, ectopic single vitrin⁺ cells were present around the pharynx and at the anterior end of the typical FoxA⁺ zone (n = 20/20 animals), even reaching the head region in some cases (Figure 3A). These ectopic cells occupied variable AP and ML locations between the original pharynx and the brain (Figure 3A, Figure 3—figure supplement 1A). This variable placement of ectopic pharyngeal cells on the ML axis was reminiscent of the patterning defect observed for eyes following map3k1 RNAi, described above. Scattered ectopic foci of cells expressing mhc-1 (a gene expressed in pharyngeal muscle) were present in the anterior half of the animal, and frequently near ectopic vitrin⁺ pharyngeal cells after 3 weeks of RNAi (Figure 3B). Clusters of pharyngeal cell types were present between the cephalic ganglia and in the pre-pharyngeal region by 6–8 weeks of map3k1 RNAi (Figure 3—figure supplement 1B).

The planarian mouth is an epidermal opening at the posterior end of the pharynx. Following 3 weeks of map3k1 RNAi, ectopic mouth cells marked by NB.22.1e appeared anterior to the typical mouth location as an anterior streak stemming from the original mouth. Rarely (n = 3/18 animals), an ectopic focus of NB.22.1e⁺ mouth cells was observed in the tail, posterior to the pharynx (Figure 3—figure supplement 1C). After 8 weeks of RNAi, ectopic scattered mouth cells were present lateral to the midline (Figure 3C). Because differentiated pharynx cells were observed outside of the canonical pharynx progenitor-specification zone, we considered the possibility that pharynx progenitor specification itself occurred in ectopic locations following map3k1 RNAi. dd_554 transcripts mark a post-mitotic pharynx progenitor population (Zhu et al., 2015). Ectopic dd_554⁺ cells were present anterior to the pharynx in map3k1 RNAi animals, including in foci (Figure 3D; Figure 3—figure supplement 1D). FoxA⁺; smedwi-1⁺ cells prominently, but not necessarily exclusively, include pharynx progenitors. These cells are predominantly centrally restricted on the AP axis to the trunk (Adler et al., 2014; Scimone et al., 2014a). We observed a small frequency of FoxA⁺; smedwi-1⁺ cells near the brain in both control and map3k1 RNAi animals (Figure 3E, Figure 3—figure supplement 1E, F). These findings are consistent with the possibility that these FoxA⁺; smedwi-1⁺ cells include pharynx progenitors, and that ectopic differentiation of pharynx progenitors can occur outside of the predominant pharynx progenitor specification zone at a low frequency.

Complete posterior duplication of the pharynx has previously been observed following ptk7, ndl-3, and wntP-2 RNAi (Scimone et al., 2016; Hill and Petersen, 2018), anterior duplication has been observed after roboA RNAi (Cebrià et al., 2007), and lateral duplication has been observed after wnt5 RNAi (Gurley et al., 2010). The map3k1 RNAi phenotype differs from these other patterning phenotypes, in that it involves greater disorganization, with the appearance of small clusters of pharyngeal cells and even single pharyngeal cells instead of only well-organized but ectopically placed pharynges. This scenario bears similarity to the phenotype for the eye: disorganized ectopic tissue differentiation, including in small clusters and single cells. These findings are consistent with the possibility that map3k1 RNAi disrupts the regulation of progenitor targeting that normally results in differentiation being restricted to occur in precise locations.

Post-mitotic progenitors normally migrate to, and differentiate at, their TZ. Prior work indicates that in addition to the TZ, the target tissue itself can incorporate its fate-specified progenitors and promote progenitor differentiation, even at an ectopic location (Atabay et al., 2018; Hill and Petersen, 2018). For instance, surgically transplanting an ectopic eye outside of the TZ, but within the broad distribution of eye progenitors, results in a stable ectopic eye that incorporates progenitors to replace dying cells as part of turnover (Atabay et al., 2018). This is enabled by the fact that progenitors in wild-type animals are specified in broad regions, giving ectopic differentiated tissues access to a constant supply of progenitors (Lapan and Reddien, 2011; Lapan and Reddien, 2012). Thus, at least two system components appear to be capable of promoting progenitor differentiation: the TZ and the target tissue. The pharynx shows similar properties to the eye – with an ectopic pharynx being maintained through progenitor incorporation and differentiation (Hill and Petersen, 2018). An ectopic organ, however, will not regenerate upon its removal. Because the original TZ location of the tissue is unchanged in this situation, progenitors will target the correct location after resection of an ectopic organ (Atabay et al., 2018; Hill and Petersen, 2018).

Given the above reasoning, ectopic eyes in map3k1 RNAi animals could, in principle, be explained by TZ movement or expansion. To assess whether TZ alteration occurred following map3k1 RNAi, or some other explanation for the map3k1 RNAi phenotype is more likely, we resected all visible eyes in map3k1 RNAi animals to determine the location of new progenitor differentiation in the absence of a target tissue. Eye-resected map3k1 RNAi animals regenerated eyes at a comparable rate to control animals, and in the normal TZ location (Figure 4A; Figure 4—figure supplement 1A). Notably, however, map3k1 RNAi animals did not regenerate any resected ectopic eyes (Figure 4A). This indicates that the TZ location is maintained after map3k1 RNAi, and that at least some progenitors are capable of reaching the normal TZ location in the absence of an eye. These data are consistent with a model in which map3k1 does not primarily control the positional information read by progenitors, but instead affects the ability of progenitors to differentiate at proper locations in response to a normal positional information system – a possibility explored further below.

To determine if the pharynx TZ is also maintained in its normal, wild-type location in map3k1 RNAi animals, we removed the entire pharynx of map3k1 RNAi and control animals (Figure 4B, Figure 4—figure supplement 1B). By 10 days after pharynx resection, animals regenerated a pharynx in the original position, but with some disorganization and multiple pharyngeal structures forming in some cases (Figure 4B, Figure 4—figure supplement 1B, C). Thus, the case for the pharynx is more complex than for the eye. Regardless, these findings for the pharynx suggest that at least the normal central TZ remains and is not simply shifted anteriorly.

To determine the location of progenitor targeting and de novo organ formation during regeneration, we amputated map3k1 RNAi animals into head, trunk, and tail fragments and analyzed them after 10 days of regeneration (Figure 4C). All fragments regenerated organs that they did not already contain at the time of amputation (Figure 4D). Two eyes formed in approximately normal locations in map3k1 RNAi head blastemas, rather than being posteriorly shifted or appearing in multiple locations initially. This suggests, like the results from the eye-resection experiments described above, that the TZ is regenerated in roughly the wild-type location in map3k1 RNAi animals.

Prior work in the planarian species D. japonica showed that map3k1 RNAi results in tail fragments regenerating pharynges in an anterior-shifted location (Hosoda et al., 2018). Consistent with this previously reported effect, pharynges regenerated more anteriorly in map3k1 RNAi S. mediterranea tail fragments, just posterior to the regenerating brain (Figure 4C–E). Notably, trunks also regenerated secondary pharynx-like aggregates very close to the brain that were underdeveloped and that appeared to interfere with structures around them (Figure 4D). Head fragments regenerated multiple pharyngeal structures (Figure 4—figure supplement 1D). Therefore, the situation for the pharynx in regeneration is more complex than that of the eye, similar to the findings for organ resections described above.

As noted above, numerous PCGs can be inhibited to cause organ duplications. However, the map3k1 RNAi phenotype described so far is largely consistent with alteration of progenitor targeting behavior rather than global shifting of positional information. To directly examine the spatial maintenance of positional information in map3k1 RNAi animals, we labeled these animals with RNA probes for multiple PCGs. Anterior, posterior, and medial PCG expression domains were largely unaffected in map3k1 RNAi animals – including expression domains for sFRP-1, ndl-4, ndl-5, ndl-2, ndl-3, wntP-2, axinB, sp5, ptk7, wnt11-1, and slit (Figure 5A). The spatial distribution of axinB transcription – a read-out of the posterior-to-anterior Wnt activity gradient in planarians – was similar in control and map3k1 RNAi animals, indicating that Wnt activity was maintained regionally on the AP axis (Figure 5A; Iglesias et al., 2011; Reuter et al., 2015; Stückemann et al., 2017; Tewari et al., 2019).

The notum⁺ anterior pole, which is produced from neoblast-derived and migratory FoxD⁺ progenitors (Roberts-Galbraith and Newmark, 2013; Scimone et al., 2014b; Vogg et al., 2014), was present, but displayed some dispersal of cells after map3k1 RNAi (Figure 5B; Figure 5—figure supplement 1A). notum⁺; chat⁺ brain cells in the cephalic ganglia also appeared more dispersed in map3k1 RNAi animals, and this pattern was somewhat reminiscent of the eye and pharynx phenotypes. notum⁺ cell disorganization was also exaggerated during tail regeneration, when the animal is challenged with generating a new head (Figure 5—figure supplement 1B). Despite some anterior pole dispersal, the posterior pole appeared mostly normal and regenerating tails were still able to obtain proper PCG expression domains by day 4 (Figure 5B; Figure 5—figure supplement 1C).

Ectopic tissues in map3k1 RNAi animals formed outside of the PCG expression domains that they are normally restricted to; for instance, opsin⁺ and dd_17258⁺ neurons formed outside of the ndl-5⁺ domain and vitrin⁺ pharyngeal cells formed outside of the wntP-2⁺ domain (Figure 5C). Ectopic differentiated cells were also observed inside PCG expression domains that they are usually not found within; for instance, opsin⁺ and dd_17258⁺ neurons formed within a domain expressing wntP-2 and wnt11-1, and vitrin⁺ cells formed within ndl-2 and ndl-5 expression domains (Figure 5C). In map3k1 RNAi tail fragments, pharynges regenerated partially outside of the normal wntP-2⁺ expression domain, despite wntP-2 having a known role in establishing trunk identity (Figure 5—figure supplement 1D). These findings are consistent with the results described above: eyes are maintained and regenerated at the normal TZ, indicating that the defining positional information for the TZ remains at approximately the correct location in map3k1 RNAi animals. The fact that ectopic eyes emerge over time in a disordered fashion suggests that it is progenitor targeting for differentiation at precise locations that is affected by map3k1 RNAi rather than positional information itself.

Recent scRNA-seq work has annotated fate-associated clusters of planarian neoblasts and post-mitotic progenitors (King et al., 2024). If map3k1 acts in progenitors to regulate their differentiation, it should be transcribed in these cells. Indeed, map3k1 transcripts were present broadly across neoblast and post-mitotic progenitor clusters (Figure 5—figure supplement 2A, B), including for the eye, neural classes affected by map3k1 RNAi, and the pharynx (Figure 5—figure supplement 2C–I). Whereas expression data alone do not necessarily indicate the site of action of map3k1, the data are consistent with the possibility that map3k1 can act within migratory progenitors.

Eye and pharynx progenitors are normally specified in restricted domains (Lapan and Reddien, 2011; Lapan and Reddien, 2012; Adler et al., 2014; Scimone et al., 2014a; Atabay et al., 2018), referred to here as progenitor specification zones. Within these domains, progenitors can migrate to reach their target tissue at their TZ, where they differentiate (Atabay et al., 2018; Hill and Petersen, 2018). We sought to test the possibility that ectopic cell differentiation, in map3k1 RNAi animals, is a result of premature progenitor differentiation at positions along the normal migratory path before reaching the TZ. An alternative scenario we considered is that ectopic progenitor specification at some distant location occurs after map3k1 RNAi and requires excessively long-range progenitor migration to reach the TZ, ultimately resulting in ectopic differentiation.

Neoblasts can be killed with irradiation (Bardeen and Baetjer, 1904), and lead shielding can be utilized in X-irradiation experiments to locally protect neoblasts, resulting in neoblasts being present only in a restricted field (Dubois, 1948; Guedelhoefer and Sánchez Alvarado, 2012; Abnave et al., 2017). Neoblasts expand slowly from these restricted regions (barring amputation) (Saló and Baguñà, 1985), but their postmitotic descendant cells serve as precursors that can readily migrate to target tissues (Guedelhoefer and Sánchez Alvarado, 2012; Abnave et al., 2017; Park et al., 2023). We utilized lead shielding over the top half of the animal to preserve the region of eye neoblasts, the place where eye progenitors are normally born, from X-irradiation (Figure 6A). Forty-eight hours after irradiation, we initiated RNAi of map3k1 to observe the behavior of eye progenitors – restricted to be born within their normal, primary eye-progenitor specification zone (the zone of surviving neoblasts). Twelve to fourteen days after the initiation of map3k1 RNAi in these partially irradiated animals, ectopic eye cells and ectopic dd_17258⁺neurons were apparent (Figure 6A, Figure 6—figure supplement 1A, B). This indicates that ectopic differentiation can occur after map3k1 RNAi even from neoblasts restricted to undergo fate specification in the normal location.

We next sought to confirm that the appearance of ectopic cells required the local production of new progenitors from neoblasts. We utilized lead shielding over the posterior half of wild-type animals and performed X-irradiation to preserve neoblasts only in the tail region (Figure 6B). We then performed map3k1 RNAi over a 10- to 12-day period and assessed animals for ectopic cell differentiation. The AP location of ectopic cells in these animals was most frequently in the area of remaining neoblasts (labeled with a probe to smedwi-1 transcript), with a small frequency of ectopic differentiation events occurring outside of this area, but still proximal to it (Figure 6B).

To assess the behavior of map3k1 RNAi eye progenitors in the normal eye-progenitor specification zone and in a non-RNAi host environment, we labeled neoblasts with EdU after 2–3 weeks of map3k1 RNAi, then transplanted a small pre-pharyngeal tissue fragment from these animals into the pre-pharyngeal region of unlabeled, wild-type host animals (hosts had not experienced map3k1 RNAi) (Figure 6C). Transplant recipient animals were then fixed 12 days later. EdU-positive cells migrated out of the transplantation region and differentiated into eye cells at both the normal eye location (at the TZ) and at ectopic locations in the host environment (Figure 6C, D; Figure 6—figure supplement 1C). Additional ectopic eye cells were observed following transplantation that were not EdU-positive, likely because of incomplete labeling or because they were born multiple divisions after the EdU pulse was delivered (Figure 6—figure supplement 1D). These observations indicate that eye progenitors originating from the normal specification zone can erroneously differentiate before reaching their target location, including in a non-RNAi host environment.

We administered map3k1 dsRNA feedings for RNAi and let animals undergo normal long-term tissue turnover to observe the consequences of errors in progenitor targeting on tissue pattern. Eight weeks of map3k1 RNAi led to differentiated pharynx muscle (mhc-1) cells within the cephalic ganglia (Figure 7A; Figure 7—figure supplement 1A). Single epidermal (NB.22.1e) cells were also observed, at a low frequency (n = 4/18), within eyes after 8 weeks of map3k1 RNAi. Tissue incorporation errors did not require many RNAi feedings to manifest. After only 3 weeks of map3k1 RNAi and tail amputation, day 10 tail regenerates (n = 9/24) displayed ectopic vitrin⁺ pharyngeal cells in eyes (Figure 7B; Figure 7—figure supplement 1B). These striking cell-organization defects were not previously observed in the patterning phenotypes of PCG RNAi animals. We suggest that this defect highlights the risk to tissue architecture of not enacting tight spatial regulation of differentiation, especially in a biological context where progenitors are spatially dispersed and migratory. Ectopic differentiated cells can become incorporated into inappropriate tissue environments where they would normally never be observed, potentially disrupting tissue structure and function.

All map3k1 RNAi animals ultimately developed tissue growths within 8–12 weeks, simply from undergoing tissue turnover without injury (Figure 7C; Figure 7—figure supplement 1C). These growths predominantly formed in the anterior of the animal and presented as a heterogeneous collection of cell and tissue types and consistently contained clusters of neurons and muscle cells (Figure 7D; Figure 7—figure supplement 1D). Pharynx (vitrin⁺) cells and gland cells (mag-1⁺) were not as common in lateral outgrowths (Figure 7D). Cell types that were present in these growths included eye cells, neurons from the central nervous system (cintillo⁺, dd_17258⁺, and dd_3524⁺ cells), glia, muscle cells, and epidermal cells (Figure 7D). We considered these aberrant growths with a heterogeneous collection of intermingled tissues to be teratomas. A similar teratoma formation defect has been observed in planarians with a defect in progenitor migration caused by integrin RNAi (Bonar and Petersen, 2017; Seebeck et al., 2017). It is known that differentiated tissues, such as the eye, can trap their own progenitors and lead to their differentiation (Atabay et al., 2018; Hill and Petersen, 2018). This suggests that ectopic differentiation in inappropriate locations in map3k1 RNAi animals can result in the trapping of additional progenitors and can ultimately lead to an inappropriate aggregate of differentiated cells with aberrant pattern and organization. This defect highlights a further risk to tissue architecture if targeting and differentiation of migratory progenitors is not tightly controlled.

Asexual S. mediterranea clonal strain CIW4 was used for all experiments. Animals were cultured in static 1x Montjuic water (1.6 mmol/l NaCl, 1.0 mmol/l CaCl₂, 1.0 mmol/l MgSO₄, 0.1 mmol/l MgCl₂, 0.1 mmol/l KCl, and 1.2 mmol/l NaHCO₃ prepared in Milli-Q water) at 20°C. Amputations were performed under cold conditions (~4°C) with a scalpel. Animals were fed homogenized beef liver weekly, with water changed biweekly. Animals were starved for approximately 7 days before experiments.

A cDNA library was generated with RNA isolated from whole animals. All genes were amplified with gene-specific primers containing adaptors for 5′ and 3′ primer regions. Amplicons were ligated into a pGEM-T Easy backbone using the pGEM-T Easy Vector Systems kit (Promega). Plasmids were transformed into NEB 5-alpha Competent E. coli (High Efficiency; New England Biolabs) and miniprepped.

Riboprobe System Components and Buffers (Promega) were used for in vitro transcription of dsRNA and RNA probes using T7 polymerase. RNA probes were transcribed using DIG, FITC, or DNP-modified nucleotides, allowing for signal amplification with conjugated DIG, FITC, and DNP antibodies. dsRNA was resuspended in water and RNA probes were resuspended in deionized formamide.

Animal mucus was removed using 5% N-acetylcysteine in PBS; animals were then fixed with 4% formaldehyde in PBST for 20 min, with rocking. Animals were then washed with PBST, incubated in 1:1 PBST: methanol, then stored in 100% methanol at −20°C until ready for bleaching. Animals were moved into mesh baskets in a 24-well plate where all remaining steps were carried out. Animals were placed on a light source to bleach for 1.5 hr in a bleaching solution (5% formamide, 0.5x SSC, and 1.2% hydrogen peroxide). After two PBST washes, animals were then treated with 5 mg/ml Proteinase K for 10 min, followed by 4% formaldehyde post-fixation in PBST.

Probes were diluted in Hybe solution (1:800) (50% deionized formamide, 5x SSC, 1  mg/ml yeast RNA, 1% Tween-20, 5% dextran sulfate), and left to incubate overnight. The following days, we performed antibody incubations at 4°C overnight using anti-DIG-POD 1:1500, Roche; 10% western blocking solution (Roche) anti-FITC-POD (1:2000, Roche; 5% horse serum, 5% western blocking solution), and anti-DNP-HRP (1:100, PerkinElmer; blocking solution with 10% inactivated Horse Serum). Tyramide signal amplification involved incubating in rhodamine (1:1000), fluorescein (1:1500), or Cy5 (1:300) in borate buffer (0.1 M boric acid, 2 M NaCl, pH 8.5) containing 0.0003% hydrogen peroxide for 10 min. Samples were incubated in 1% sodium azide for 2 hr to inactivate the HRP. Blocking and antibody incubations then occurred for detection of the second probe. Animals were incubated overnight in 1 mg/ml DAPI solution at 4°C. Animals were mounted on coverslips in ProLong Gold Antifade Mountant (Thermo Fisher).

F-ara-EdU (Click Chemistry Tools) was diluted in Dimethyl sulfoxide (DMSO) to 200 mg/ml, then diluted in static 1x Montjuic water to 1.25 mg/ml. Animals were split into 10 animals per well in a 12-well plate, then soaked in 1.25 mg/ml EdU solution for 20 hr following 1 week of starvation. EdU solution was replaced with 5 mg/ml Instant Ocean Sea salt dissolved in Milli-Q water. Prior to probe hybridization in the in situ hybridization protocol, following proteinase K and 4% formaldehyde incubations, cells were incubated in a ‘click reaction’ – 1% 100 mM CuSO₄, 0.1% 10 mM TAMRA-Azide-fluor 545 (Sigma-Aldrich), and 20% 50 mM ascorbic acid in PBS for 30 min in the dark, proceeded by six PBST washes and continuation of the probe hybridization step.

C. elegans unc-22 dsRNA was used as the negative control for all RNAi experiments. 50 µl of homogenized beef liver was mixed with 25 µl of dsRNA prep and 3 µl of a 1:1 mixture of MilliQ water and red food coloring. map3k1 RNAi experiment durations ranged from 1 to 16 weeks of dsRNA feedings for RNAi. Time courses were conducted using animals fed 1, 2, and 3–4 weeks of dsRNA. Animals used to study outgrowth phenotypes were fixed between 8 and 12 weeks of RNAi, based on teratoma severity. Animals fed dsRNA for 3 and 4 weeks showed similar phenotype severity and were often analyzed as one group. Animals were given food for 1 hr, twice a week, for the first 8 feedings (4 weeks); animals were then given food once a week, for 30 min, for all subsequent feedings to prevent rapid growth and fissioning (separation of the tail from the body).

EdU plug transplants were performed using map3k1 or control RNAi animals, 12 hr following a 20-hr EdU pulse, as the donor to a recipient wild-type animal. Donor animals were anesthetized with 0.2% chlorotone solution, followed by an incubation in Holfreter’s solution, then placed on an ice block covered in Whatman filter paper moistened with 1x Montjuic water to surgically manipulate with a clean scalpel. EdU-positive map3k1 RNAi donor animals and wild-type recipients both had a center portion of their pre-pharyngeal regions removed. The pre-pharyngeal donor graft from the map3k1 RNAi animal was placed in the EdU-negative wild-type recipient’s pre-pharyngeal region. Recipients were then gently covered with cigarette paper soaked in chilled Holtfreter’s solution and transferred to a small Petri dish with enough Holtfreter’s solution to cover the bottom of the dish. Petri dishes were put at 10°C for 20 hr; the following day, transplant recipients were gently recovered and put into 1x Montjuic water containing 0.1% gentamicin (Gibco) to heal. Water was changed every 2 days, and animals were fixed at day 12 post-transplant.

Animals were irradiated using an X-Rad320, Precision X-Ray Irradiation chamber. For shielded irradiation experiments, animals were anesthetized with 0.2% chlorotone, then arranged on Whatman filter paper in a Petri dish sitting on ice. Animals were oriented to have their anterior half covered by the lead shield placed over the Petri dish. Samples were placed in the irradiation chamber and exposed to 3000 Rad of unidirectional X-irradiation. Animals were rescued with 1x Montjuic water and stored in 1x Montjuic water with 0.1% gentamicin (Gibco) to recover. Water was changed every 2 days. RNAi experiments were carried out starting 2 days after irradiation exposure. Anterior half-shielded animals were fed dsRNA for a period of 12–14 days before fixation, and posterior half-shielded animals were fed dsRNA for a period of 10–12 days before fixation; these ranges were dependent on the health of the animals at the time of fixation. If animals started showing slight signs of health decline (e.g., small lesions, slight head regression, bloating), they were fixed on that day.

All tissue resections and amputations were performed by placing animals on wet filter paper on top of a cold block to minimize movements during surgeries. Animals were kept moist throughout all procedures with 1x Montjuic water. Eyes were resected using a small scalpel. Curved edges were created by lightly tapping the tip of the scalpel on a clean, hard surface. The curved edge was used to poke and scoop out the eye in a poke and pull motion. Pharynges were resected by puncturing a diamond shape around the pharynx with a small scalpel, then gently removing the pharynx tissue. Animals were placed in 1x Montjuic water with 0.1% gentamicin (Gibco) to recover, and water was changed every 2 days.

FISH images were analyzed using Fiji Software. The AP axis was binned into six regions according to anatomical landmarks: AP_1 (head tip → bottom of the brain), AP_2 (bottom of the brain → top of the pharynx), AP_3 (top of the pharynx → middle of the pharynx), AP_4 (middle of the pharynx → bottom of the pharynx), AP_5 (bottom of the pharynx → halfway between the bottom of the pharynx and the tail tip), and AP_6 (halfway between the bottom of the pharynx and the tail tip → tail tip). Each animal had one data point in each of the six AP bins. map3k1 and control RNAi ectopic PRN and OC cell counts at 3–4 weeks were analyzed by generating a Poisson generalized linear mixed model, using AP bin as a random covariate. A Mann–Whitney U test was used for dd_17258⁺ and EdU⁺ cell counts in transplants to account for non-normal data distributions. Any tail or head cell count data sets showing overdispersion – gland cells (dd_7131 and dd_8476) and FoxA⁺; smedwi-1⁻ cell counts between the brain – were analyzed with a negative binomial regression, correcting for overdispersion. Two-tailed permutation tests, using 10,000 permutations per test, were carried out for ovo⁺ cell counts in the tail because of the low sample number and zero-inflated dataset. A binomial exact test was carried out on ectopic cells inside versus outside the smedwi-1⁺ zone after shielded irradiation and map3k1 RNAi, assuming 50% probability of either outcome. Prism software was used to carry out Student’s t-tests, Mann–Whitney U tests, and binomial exact tests. R Studio was used to compute negative binomial regression, Poisson regression, and permutation tests.