kotahi

Metadata

eLife Assessment

This study investigates whether heavy metal stress can induce maize-like phenotypic and molecular responses in teosinte and whether these responses overlap with genomic regions implicated in domestication. By combining copper and cadmium treatments with quantitative phenotyping, gene-expression analyses, and expanded assessments of nucleotide diversity across a key chromosome 5 interval, the authors provide an integrated view of how abiotic stress responses intersect with domestication-related traits. The significance of the findings is valuable, as the work offers meaningful insights for the subfield of maize evolution and stress biology by extending heavy-metal response analyses to teosinte and linking them to domestication-associated loci, although the evolutionary implications remain indirect. The strength of evidence is solid, with appropriately designed and quantitatively supported experiments that broadly support the claims, but do not yet establish a causal or historical role for heavy metal stress in domestication.

Reviewer #1 (Public review):

In this study, Acosta-Bayona et al. investigate whether heavy metal (HM) stress can induce phenotypic and molecular responses in teosinte parviglumis that resemble traits associated with domestication, and whether genes within a domestication-linked region show patterns consistent with reduced genetic diversity and signatures of selection. The authors exposed both maize and teosinte parviglumis to a fixed dose of copper and cadmium, representing an essential and a non-essential element, respectively. They assessed shoot and root phenotypic traits at a defined developmental stage in plants exposed to HM stress versus control. They then integrated these phenotypic results with expanded analyses of genetic diversity across a broader chromosome 5 interval, which was previously associated with domestication-related traits. Overall, the revisions improve the clarity and the robustness of the analyses, as well as make the conclusions better aligned with the evidence.

The revised manuscript is strengthened by several additions.

(1) The authors broaden the genetic analysis beyond a small set of loci and evaluate nucleotide variability across several genes within the linked chromosome 5 interval, which improves the interpretation of diversity patterns and reduces concerns about a too narrow locus selection or regional linkage effects driving the conclusions.

(2) The expression analyses are now presented with clearer methodological separation and stronger quantitative support. Now, tissue/developmental RT-PCR profiles are distinguished from real-time qPCR assays used to test HM-induced expression changes, with appropriate replication and statistical reporting.

(3) The authors include a transcriptome-scale element by analyzing multiple published and publicly available HM-stress transcriptome datasets and reporting shared differentially expressed genes across studies, which supports the interpretation that the observed expression changes align with broader HM-responsive transcriptional programs.

However, it remains challenging to distinguish which aspects of the HM responses observed here represent novel insight versus patterns already reported in maize HM-stress studies. In addition, the link between HM exposure and domestication history remains indirect: reduced diversity patterns and stress-responsive expression do not, on their own, demonstrate human-driven selection or a specific paleoenvironmental scenario, and alternative explanations related to general stress responses or regional evolutionary processes cannot be fully excluded.

Reviewer #2 (Public review):

Summary:

This work explores the phenotypic developmental traits associated with Cu and Cd responses in teosinte parviglumis, a species evolutionary related to extant maize crops. Cu and Cd could serve as a proxy for heavy metals present in the soils. The manuscript explores potential genetic loci associated with heavy metal responses and domestication. This includes heavy metal transporters which are unregulated during stress. To study that, authors compare the plant architecture of maize defective in ZmHMA1 and speculate on the association of heavy metals with domestication.

Strengths:

Very few studies covered the responses of teosintes to heavy metal stress. The physiological function of ZmHMA1 in maize is also valuable. The idea and speculation section is interesting and well-implemented.

Weaknesses:

Some conclusions are still speculative and future experiment could provide more clues about potential molecular mechanisms for the ideas proposed here.

Author response:

The following is the authors’ response to the original reviews

Public Reviews:
Reviewer #1(Public review):
In this study, Acosta-Bayona et al. aim to better understand how environmental conditions could have influenced specific gene functions that may have been selected for during the domestication of teosinte parviglumis into domesticated maize. The authors are particularly interested in identifying the initial phenotypic changes that led to the original divergence of these two subspecies. They selected heavy metal (HM) stress as the condition to investigate. While the justification for this choice remains speculative, paleoenvironmental data would add value; the authors hypothesize that volcanic activity near the region of origin could have played a role.

The justification of choice to investigate the effects of heavy metal stress is not speculative. As mentioned now in the Abstract, the elucidation of the genome from the Palomero toluqueño maize landrace revealed heavy metal effects during domestication (Vielle-Calzada et al., Science 2009). Our aim was to test the hypothesis that heavy metal (HM) stress influenced the evolutionary transition of teosinte parviglumis to maize.

(1) Although the paper presents some interesting findings, it is difficult to distinguish which observations are novel versus already known in the literature regarding maize HM stress responses. The rationale behind focusing on specific loci is often lacking. For example, a statistically significant region identified via LOD score on chromosome 5 contains over 50 genes, yet the authors focus on three known HM-related genes without discussing others in the region. It is unclear why ZmHMA1 was selected for mutagenesis over ZmHMA7 or ZmSKUs5.

We appreciated the depth and value of this comment.

Maize phenotypic responses to sublethal concentrations to heavy metals – copper (Cu) and cadmium (Cd) in particular - are well characterized and published, and in agreement with our results. In the first section of the Results (pgs 7 and 8), we added pertinent references to clearly show which observations are already known. By contrast, teosinte parviglumis responses are in all cases novel. To our knowledge this is the first study that analyzed in detail the phenotypic response of teosinte to sublethal concentrations of heavy metals, specifically Cu and Cd. We have now emphasized the novelty of these observations (pg 8).

To address the fact that we only focused on three known HM-related genes without discussing others in the statistically significant region identified via LOD score on chr.5, we have added a full section that reads as follows (pgs. 11 to 13 of the new version):

“Large-scale genomic and transcriptomic comparisons indicate that many HM response genes were positively selected across the maize genome.

To expand the results well beyond the analysis of the three genes previously described, we performed a detailed analysis of genetic diversity across the 11.47 Mb genomic region comprised between ZmSKUs5 and ZmHMA1. This additional analysis reveals general tendencies in the quantity and nature of loci that were affected by positive selection during the teosinte parviglumis to maize transition in a region identified via LOD score on chr.5. We compared nucleotide variability by using 100 bp bins covering loci composed of two 30 Kb segments up and downstream of coding sequences, respectively, and the coding sequence itself, for 173 genes present within the genomic region comprised between ZmSKUs5 and ZmHMA (Figure S1 and Supplementary File 6). Two types of statistical tests (ANOVA and Wilcoxon) were applied to nucleotide variability comparisons using the entirety of each locus. The Benjamini-Hochber procedure allowed an estimation of the false discovery rate (FDR<0.05) to avoid type I errors (false positives). Although some individual loci appear as differently classified depending on the statistical test applied (22 out of 173 loci), the general differences in nucleotide variability are consistently maintained within the subregions described below. We found that 166 out of 173 loci show signatures of positive selection and are roughly organized in five independent subregions of variable length. The first six loci are consecutively ordered in a 402 Kb subregion that includes ZmSKUs5. A second group of 13 consecutive loci expands over a 1.44 Mb subregion that contains NRAMP ALUMINUM TRANSPORTER1, also involved in HM response through uptake of divalent ions. A third group of 17 consecutive loci expands over 1.28 Mb; eleven contain genes encoding for uncharacterized proteins. The fourth group is composed of 57 consecutive loci expanding over 3.22 Mb and contains genes encoding for DEFECTIVE KERNEL55, AUXIN RESPONSE FACTOR16, and peroxydases involved in responses to oxydative stress. The fifth group contains 12 consecutive loci expanding over 713 Kb and contains ZmHMA1. An additional segment of approximately 1.17 Mb and containing 25 consecutive loci that were positively selected expands away from the ZmSKUs5-ZmHMA1 segment; it also contains several genes encoding for peroxydases. Although multiple loci include genes that could be involved in abiotic stress and oxidative responses, these results suggest that multiple factors other than HM stress could have played a role in the evolutionary mechanisms that affected the genetic diversity of chr.5 during the teosinte parviglumis to maize transition.

To further analyze the possibility that HM response could have played a role in maize emergence and subsequent domestication, we analyzed large scale transcriptomic data corresponding to independent experiments aiming at understanding the response of maize roots to HM stress. Six available transcriptomes were selected for in-depth analysis because they presented a fold change strictly higher than 1, and their results were supported by false discovery rates (FDR<0.05). These six transcriptomes (Table S5) included HM response datasets corresponding to growth conditions that not only incorporated Cu, but also lead (Pb) and chromium (Cr) that were not included in the substrate of our experiments. Transcriptional profiles were obtained from roots of plants at different stages: maize seedlings (Shen et al., 2012; Gao et al., 2015; Zhang et al., 2024a), three week old plantlets (Yang et al., 2023), and plants at V2 stage (Zhang et al., 2024b; Fengxia et al., 2025). A total of 120 genes shared by all six transcriptomes were found to be differentially expressed under HM stress conditions (66 upegulated and 54 downregulated; Figure S3), including ZmSKUs5, ZmHMA1 and ZmHMA7; 52 of them (43.3%) are located in maize loci showing less than 70% of the nucleotide variability found in teosinte parviglumis, suggesting that they were affected by positive selection (Yamasaki et al., 2005; Supplementary File 7). Of 18 mapping in chr.5, twelve are within the 82 cM that fractionates into multiple QTLs under selection during the parviglumis to maize transition. Interestingly, five additional loci containing HM response genes completely lack SNPs within their total length in both parviglumis and maize, and 19 additional loci lack SNPs in at least one 30 Kb segment or their coding region (Supplementary File 7), suggesting the frequent presence of ultraconserved genomic regions in many loci containing HM response genes. When this same analysis was conducted in a set of loci comprising 63 genes previously identified as differentially expressed in response to abiotic stress not directly related to HM responses (hypoxia; nutritional deficiency; soil alkalinity; drought; soil salinity), 18 loci (28.6%) showed less than 70% of the nucleotide variability found in teosinte parviglumis. Only one of them maps in chr.5 and none contained segments or coding regions lacking SNPs in parviglumis or maize. These results suggest that in contrast to other types of abiotic stress response genes, loci comprising a large set of genes that unambiguously respond to HM stress caused by chemical elements of diverse nature were affected by positive selection during the parviglumis to maize transition, irrespectively of their position in the genome.”

The detailed analysis of genetic diversity across 11.47 Mb of chr.5 in the genomic region comprised between ZmSKUs5 and ZmHMA1 in presented as Supplementary File 6.

The analysis of genetic diversity in loci encompassing heavy metal response genes shared by six transcriptomes and abiotic stress controls are described in Supplementary File 7.

In the Discussion (pgs. 21 and 22), we added a paragraph section that reads as follows:

“Although loss of genetic diversity is usually the result of human selection during domestication, it can also represent a consequence of natural selective pressures favoring fitness of specific teosinte parviglumis allelic variants better adapted to environmental changes and subsequently affected by human selection during the domestication process. This possibility is reflected by widely spread selective sweeps affecting a large portion of chr.5 that contains hundreds of genes showing signatures of positive selection. The analysis of 11.47 Mb covering the ZmHMA1ZmSKUs5 segment confirms the presence of large but discrete genomic subregions that were positively selected during the teosinte parviglumis to maize transition. Although several contain genes involved in HM response and oxidative stress, the diversity of gene functions does not necessarily favor abiotic stress over other factors that could be at the origin of selective forces affecting these regions. By contrast, a large scale transcriptomic survey indicates that genes consistently responding to HMs (Cu, Cd, Pb and Cr ) show signatures of positive selection at unusual high frequencies (43.3%) as compared to loci containing genes responding to other types of abiotic stress (28.6%). Our identification of HM response genes affected by positive selection is far from being exhaustive. Nevertheless, it agrees with the expected effects of a widespread selective sweep caused by environmental changes that influenced the parviglumis to maize transition at the genetic level. Of intriguing interest are 24 loci that partially or completely lack SNPs in both teosinte parviglumis and maize, suggesting possible genetic bottlenecks occurred before the teosinte to maize transition. Examples of other edaphological factors driving genetic divergence either in the teosintes or maize include local adaptation to phosphorus concentration in mexicana and parviglumis (Aguirre-Liguori et al. 2019), and fast maize adaptation to changing iron availability through the action of genes involved in its mobilization, uptake, and transport (Benke and Stich 2011). Our results reveal a teosinte parviglumis environmental plasticity that could be related to the function of HM response genes positively selected during the teosinte parviglumis to maize transition. Previous studies have demonstrated that transposable elements (TEs) contribute to activation of maize genes in response to abiotic stress, affecting up to 20% of the genes upregulated in response to abiotic stress, and as many as 33% of genes that are only expressed in response to stress (Makarevitch et al., 2015). It is therefore possible that the HM response of some specific genes that influenced maize emergence or domestication could be mediated by TEs influencing or driving their transcriptional regulation.”

The mutagenic analysis of ZmHMA7 and ZmSKUs5 will be included in a different publication.

(2) The idea that HM stress impacted gene function and influenced human selection during domestication is of interest. However, the data presented do not convincingly link environmental factors with human-driven selection or the paleoenvironmental context of the transition. While lower nucleotide diversity values in maize could suggest selective pressure, it is not sufficient to infer human selection and could be due to other evolutionary processes. It is also unclear whether the statistical analysis was robust enough to rule out bias from a narrow locus selection. Furthermore, the addition of paleoclimate records (Paleoenvironmental Data Sources as a starting point) or conducting ecological niche modeling or crop growth models incorporating climate and soil scenarios would strengthen the arguments.

We think that the detailed analysis of genetic diversity across 11.46 Mb covering the ZmSKUs5 to ZmHMA1 genomic segment – and its statistical validation - provides a precise understanding of the selective sweep dimensions in chr.5.

We do agree that lower nucleotide diversity values in maize are not sufficient to infer human selection. Because many HM response loci show unusually low nucleotide variability in teosinte parviglumis (see the results of the transcriptomic analysis presented above), we cannot discard the possibility that natural selection forces related to environmental changes could have affected native populations of teosinte parviglumis.

To further explore the link between environmental factors, natural or human-driven selection, and the paleoenvironmental context of the parviglumis to maize transition, we revised paleoenvironmental and geological records and added results in two sections that read as follows (pgs. 17 to 20):

“Paleoenvironmental studies reveal periods of climatic instability in the presumed region of maize emergence during the early Holocene.

It is well accepted that temperature fluctuations, volcanism and anthropogenic impact shaped the distribution and abundance of plant species in the Transmexican Volcanic Belt (TMVB) during the last 14,000 years (Torrescano-Valle et al. 2019). The TMVB has produced close to 8000 volcanic structures (Ferrari et al., 2011), transforming the relief multiple times, and causing hydrographic and soil changes that actively modified the distribution and composition of plant communities in Central Mexico. Detailed paleoenvironmental data for the Pleistocene and Holocene is available for several lacustrine zones located within the 50 to 100 km range of the region currently considered the cradle of maize domestication (Matzuoka et al. 2002; Figure 5a). In Lake Zirahuén (102°44′ W; 19°26′ N and approximately 2075 meters above sea level; index [i] in Figure 5a), pollen, microcharcoal and magnetic susceptibility analyses of two sedimentary sequences reveals three periods of major ecological change during the early and middle Holocene.

Between 9500 and 9000 calibrated years before present (cal yr BP), pine forests seem to have been associated with summer insolation increases. A second peak of forest change occurred at around 8200 cal yr BP, coinciding with cold oscillations documented in the North Atlantic. Finally, events occurred between 7500 and 7100 cal yr BP shows an abrupt change in the plant community related to humid Holocene climates and a presumed volcanic event (Lozano-García et al., 2013). The environmental history of the central Balsas watershed has also been documented by pollen, charcoal, and sedimentary analysis conducted in three lakes and a swamp of the Iguala valley (Piperno et al. 2007). Paleoecological records of lake Ixtacyola (8°20N, 99°35W and approximately 720 meters above sea level; index [ii] in Figure 5a) and lake Ixtapa (8°21N, 99°26W) indicate that an important increase in temperature and precipitation occurred between 13000 and 10000 cal yr BP. The pollen record of Ixtacyola showed that members of the genus Zea were already part of the vegetation coverage by 12900 to 13000 cal yr BP, suggesting that some teosintes – likely including parviglumis - were commonly found at elevation areas where they do not presently occur. Lake Almoloya (also named Chignahuapan; 19°05N, 99°20E and approximately 2575 meters above sea level; index [iii] in Figure 5a) in the upper Lerma basin is only 20 Km from the crater of the Nevado de Toluca that is responsible for creating the late Pleistocene Upper Toluca Pumice layer over which the Lerma basin is deposited. Pollen records indicate the presence of Zea species by 11080 to 10780 cal yr BP. As for other locations, an important period of climatic instability prevailed between 11500 and 8500 cal yr BP (Ludlow-Wiechers et al., 2005). Humidity fluctuations occurred until 8000 cal yr BP, with a stable temperate climate between 8500 and 5000 cal yr BP. Although pollen and diatom studies are often difficult to interpret at a regional scale, the overall results presented above suggest consistent periods of Zea plants present in periods of environmental and climatic instability that correlate with the history of volcanic activity during the early Holocene, as described in the next section.

Temporal and geographical convergence between volcanic eruptions and maize emergence during the Holocene.

Current evidence indicates that the emergence and domestication of maize initiated in Mesoamerica some time around 9,000 yr BP (Matsuoka et al. 2002). The current location of teosinte parviglumis populations that are phylogenetically most closely allied with maize are currently distributed in a region located between the Michoacan-Guanajuato Volcanic Field (MGVF) at their northwest, and the Nevado de Toluca and Popocatéptl volcanoes at their east and northeast (Figure 5a; Matsuoka et al. 2002). Precise records of field data indicate that ten accessions were collected in the Balsas river drainage near Teloloapan and Sierra de Huautla (Guerrero), at approximately 100 km south of the Nevado de Toluca crater. Three other accessions were collected near Tejupilco de Hidalgo and Zacazonapan (Estado de México), at approximately 50 to 60 km from the Nevado de Toluca crater (8762, JSG y LOS-161, and JSG-391). And four other accessions were located in Michoacan, at a location within the MGVF (accession 8763), or at mid-distance between the MGVF and the Nevado de Toluca crater (accessions JSG y LOS-130, 8761, and 8766).

The most important source of HMs in ancient soils of Mesoamerica is TMBV-dependent volcanic activity through short- and long-term effects related to lava deposits, ores, hydrothermal flow, and ash (Torrescano-Valle et al. 2019). The Nevado de Toluca volcano produced one of the most powerful eruptions from central Mesoamerica in the Holocene, giving rise to the Upper Toluca Pumice deposit at 12621 to 12025 cal yr BP (Arce et al., 2003; Figure 5b). The pumice fallout blanketed the Lerma and Mexico basins with 40 cm of coarse ash (Bloomfield and Valastro 1977; Arce et al. 2003). A second eruption dated by 36Cl exposure occurred at 9700 cal yr BP (Arce et al. 2003; Figure 5b), and the most recent eruption occurred at 3580 to 3831 cal yr BP (Macías et al. 1997). During the early and middle Holocene, the Popocatéptl volcano produced at least four eruptions dated 13037-12060, 10775–9564, 8328-7591, and 6262-5318 cal yr BP (Siebe et al. 1997); three other important eruptions occurred during the late Holocene, between 2713 and 733 cal yr BP (Siebe and Macías, 2006). In addition, the MGFV is a monogenetic volcanic field for which 23 independent eruptions have been documented during the Holocene, 21 of them located towards the southern part of the field, in close proximity to the region harboring some of the teosinte parviglumis populations most closely related to maize. Three of these eruptions occurred in the early Holocene (El Huanillo 1130 to 9688 cal yr BP; La Taza 10649 to 10300 cal yr BP; Cerro Grande 10173 to 9502 cal yr BP; Figure 5b), and three others during the initial period of the middle Holocene, between 8400 and 7696 cal yr BP (La Mina, Los Caballos, and Cerro Amarillo; Figure 5b). On average, a new volcano forms every ~435 years in the MGFV (Macías and Arce, 2019). No less than 16 other eruptions occurred between 7159 cal yr BP and the present time (Figure 5b). Soils of volcanic origin (andosols) are currently distributed in regions north-west from the Nevado de Toluca and Popocatéptl craters, in close proximity with teosinte parviglumis populations most closely related to maize (Figure S5). Although modern distribution of teosinte populations may differ from their distribution around 9000 yr BP, and unknown populations more closely related to maize may yet to be discovered, this data indicates that the date and region where maize emerged is convergent with the dates and locations of several volcanic eruptions occurred during the Holocene in that same region.”

(3) Despite the interest in examining HM stress in maize and the presence of a pleiotropic phenotype, the assessment of the impact of gene expression is limited. The authors rely on qPCR for two ZmHMA genes and the locus tb1, known to be associated with maize architecture. A transcriptomic analysis would be necessary to 1- strengthen the proposed connection and 2- identify other genes with linked QTLs, such as those in the short arm of chromosome 5.

Real-time qPCR is an accurate and reliable approach to assess the expression of specific genes such as ZMHMA1 and Tb1, but we agree that our results do not allow to establish a direct regulatory link between the function of Tb1, the pleiotropic parviglumis phenotype under HM stress, and the function of ZmHMA1. We also concede that the large transcriptional analysis of HM response in maize (presented above) does not allow to elucidate a possible connection between these two genes. We have substantially downplayed our conclusion in this section by modifying the end of the section in pg. 17, that now reads:

“These results do not allow to directly link the regulation of ZmHMA1 expression to the function of Tb1; however, they open an opportunity to further investigate the possibility that under HM stress, the formation of secondary ramifications in teosinte parviglumis could be repressed by transcription factors of the TCP family, including Tb1.”

This is also emphasized in the Discussion (pg 21) as follows:

“Under HM stress, we also show that Tb1 is overexpressed in the apical meristem of teosinte parviglumis, suggesting that formation of secondary ramifications is repressed by Tb1 function under HM stress, as in extant maize. At this stage we cannot discard the possibility that Tb1 upregulation in parviglumis reflects a more generalized response to abiotic stress; however, the expression ZmHMA1 is downregulated in W22 wild-type maize meristems in the presence of HMs but upregulated in teosinte parviglumis meristems, suggesting that a specific regulatory shift relating HM responses and ZmHMA1 function occurred during the teosinte parviglumis to maize transition.”

On the other hand, the transcriptional analysis the identification of 52 additional HM response genes showing signatures of positive selection occurred during the parviglumis to maize transition; 12 of them map to chr.5 within the region having linked QTLs within the short arm of chr.5. So far, genes involved in HM response and oxidative stress represent the most prevalent class of genes identified within the genomic region showing pleiotropic effects on domestication and multiple linked QTLs in chr.5.

Reviewer #2 (Public review):
Summary:
This work explores the phenotypic developmental traits associated with Cu and Cd responses in teosinte parviglumis, a species evolutionary related to extant maize crops. Cu and Cd could serve as a proxy for heavy metals present in the soils. The manuscript explores potential genetic loci associated with heavy metal responses and domestication identified in previous studies. This includes heavy metal transporters, which are unregulated during stress. To study that, the authors compare the plant architecture of maize defective in ZmHMA1 and speculate on its association with domestication.

Strengths:
Very few studies covered the responses of teosintes to heavy metal stress. The physiological function of ZmHMA1 in maize also gives some novelty in this study. The idea and speculation section is interesting and well-implemented.

Weaknesses:
The authors explored Cu/Cd stress but not a more comprehensive panel of heavy metals, making the implications of this study quite narrow. Some techniques used, such as end-point RT-PCR and qPCR, are substandard for the field. The phenotypic changes explored are not clearly connected with the potential genetic mechanisms associated with them, with the exception of nodal roots. If teosintes in response to heavy metal have phenotypic similarity with modern landraces of maize, then heavy metal stress might have been a confounding factor in the selection of maize and not a potential driving factor. Similar to the positive selection of ZmHMA1 and its phenotypic traits. In that sense, there is no clear hypothesis of what the authors are looking for in this study, and it is hard to make conclusions based on the provided results to understand its importance. The authors do not provide any clear data on the potential influence of heavy metals in the field during the domestication of maize. The potential role of Tb-1 is not very clear either.

Thank you for these comments. We have now emphasized our hypothesis in the abstract and the last paragraph of the Introduction (pg. 6):

“To test the hypothesis that heavy metal (HM) stress influenced the evolutionary transition of teosinte to maize, we exposed both subspecies to sublethal concentrations of copper and cadmium etc…”

A comprehensive panel of heavy metals would not be more accurate in terms of simulating the composition of soils evolving across 9,000 years in the region where maize presumably emerged. Copper (Cu) and cadmium (Cu) correspond each to a different affinity group for proteins of the ZmHMA family. ZmHMA1 has preferential affinity for Cu and Ag (silver), whereas ZmHMA7 has preferential affinity to Cd, Zn (zinc), Co (cobalt), and Pb (lead). Since these P1b-ATPase transporters mediate the movement of divalent cations, their function remains consistent regardless of the specific metal tested, provided it belongs to the respective affinity group. By applying sublethal concentrations of Cd (16 mg/kg) and Cu (400 mg/kg), we caused a measurable physiological response while allowing plants to complete their life cycle, including the reproductive phase, facilitating a comprehensive analysis of metal stress adaptation. Whereas higher doses impair flowering or are lethal, lower Cu/Cd concentrations do not consistently show conventional phenotypic responses such as reduced plant growth (AbdElgawad et al. 2020; Atta et al., 2023)

Based on comments by both reviewers, we present now a large transcriptional analysis that incorporates HM responses to lead (Pb) and chromium (Cr), in addition to Cu. Results show that many genes responding to Pb and Cr were also positively selected across the maize genome, suggesting that HM stress led to a ubiquitous rather than a specific evolutionary response to heavy metals (please see our response to Reviewer#1 and sections in pgs. 11 to 13) .

Real-time qPCR is an accurate and reliable approach to assess the expression of specific genes such as ZMHMA1 and Tb1, but we agree that our results do not allow to establish a direct regulatory link between the function of Tb1, the pleiotropic parviglumis phenotype under HM stress, and the function of ZmHMA1. We also concede that the large transcriptional analysis of HM response in maize (presented above) does not allow to elucidate a possible connection between these two genes. Therefore, we have substantially downplayed our conclusion in this section by modifying the end of the section in pg. 17, that now reads:

There are two phenotypic changes clearly connected with the genetic mechanisms involved in the parviglumis to maize transition: plant height and the number of seminal roots (not nodal roots). These changes have been now emphasized in the Abstract and the description of the results.

Regarding the possibility for HM stress to represent a confounding factor in the selection of maize and not a driving factor, we expanded the genomic analysis of genetic diversity well beyond the analysis of the three genes under initial study, to cover a segment of 11.47 Mb comprised between ZmSKUs5 and ZmHMA1. We compared nucleotide variability by using 100 bp bins covering loci composed of two 30 Kb segments up and downstream of coding sequences, respectively, and the coding sequence itself, for 173 genes present within the genomic region comprised between ZmSKUs5 and ZmHMA (Figure S1 and Supplementary File 6). The full analysis is presented in a new section pgs. 11 and 12. We found that 166 out of 173 loci show signatures of positive selection and are roughly organized in five independent subregions of variable length. Four out of five subregions contain more than one HM or oxidative stress response gene within loci showing signatures of positive selection. Although multiple factors other than HM stress could have played a role in the evolutionary mechanisms that affected the genetic diversity of chr.5, large scale transcriptomic data corresponding to independent experiments aiming at understanding the response of maize roots to HM stress allowed the identification of 49 additional HM response genes within loci showing positive selection across the genome, a proportion (43.3%) far greater than the proportion of loci containing response genes to other types of abiotic stress not related to HMs (28.6%). These results are described in detail in pgs. 12 and 13 (Figure S3 and Supplementary File 7). These results provide strong evidence in favor of HM stress and not another factor driving positive selection.

We now provide precise and pertinent paleoenvironmental data on the potential influence of heavy metals in the field. In sections pgs. 17 to 20 we review paleoenvironmental studies revealing periods of climatic instability in the presumed region of maize emergence during the early Holocene, and data indicating that the date and region where maize emerged is convergent with the dates and locations of several volcanic eruptions occurred during the early and middle Holocene in that same region. Please see responses to Reviewer#1 for details.

We agree that our results do not allow to establish a direct regulatory link between the function of Tb1, the pleiotropic parviglumis phenotype under HM stress, and the function of ZmHMA1. We also concede that the large transcriptional analysis of HM response in maize (presented above) does not allow to elucidate a possible connection between these two genes. Therefore, we have substantially downplayed our conclusion in this section by modifying the end of the section in pg. 17, that now reads:

This is also emphasized in the Discussion (pg 21) as follows:

Recommendations for the authors:
Reviewer #1 (Recommendations for the authors):
While the dataset generated provides an interesting foundation for hypothesis testing on HM stress and domestication, the current data do not sufficiently support the conclusions of the manuscript.
(1) The description of maize and teosinte architecture under HM stress is well presented.

However, traits like shoot height, leaf size reduction, and biomass loss also occur under other environmental stresses such as drought and salinity. Additional evidence beyond shoot and root architecture would help validate the link between tb1 expression and specific ZmHMA genes under HM stress, or whether it reflects a more generalized stress response.

We have already addressed in detail this point in the public response to Reviewer#1.

(2) The nucleotide variability analysis is interesting, but I would have liked to see additional information to clarify the choice of the data selection and the strength of the conclusions with human selection.

We have already addressed in detail this point in the public response to Reviewer#1.

a) The choice of Tripsacum dactyloides as the outgroup to determine nucleotide variability seems to be distant, and I wonder whether other combinations with a closer outgroup or multiple outgroups were tried to provide a more accurate context.

Nucleotide variability in Tripsacum dactyloides is used to graphically illustrate an external reference and not as an outgroup in the extended analysis of genetic diversity at the locus and genomic level. We did not used Tripsacum dactyloides as an outgroup in our statisticalm analysis. We could have indeed a closer teosinte subspecies as an outgroup, but at this stage no data warrants that environmentally-related selective pressures could have affected genetic diversite in other teosintes. This possibility in currently being investigated.

b) Evolutionary differences not related to human influence could affect the results. The phrase "order of magnitude difference in π values" needs statistical validation (e.g., confidence intervals, p-values).

We agree and have eliminated the sentence, as it is no longer relevant at the light of the detailed genomic analysis of genetic diversity prsented in Supplementary File 6.

c) The comparison with ZmGLB1, a neutral control locus, suggests that domestication-related changes in nucleotide variability are specific to the three candidate genes. However, the concept of neutrality is complex, and while ZmGLB1 may be considered neutral in this case, the argument does not address the possibility of other factors, such as linked selection, that could influence variability in these genes. Referencing Hufford et al. is insufficient and would require a deeper argument.

We also agree with this comment. We think that the influence and consequences of linked selection are now well documented for 11.46 Mb analyzed in chr.5 (pgs 11 and 12) in the main text and Supplementary File 6).

(3) The statement: "Our evidence indicates that HM stress revealed a teosinte parviglumis environmental plasticity that is directly related to the function of specific HM response genes that were affected by domestication through human selection" is not supported by the presented data. The rationale for the specific Cd/Cu dosage used is unclear. A dose-response gradient would better demonstrate the nature and strength of the plastic response.

Previous reports support the rationale for the specific HM dosage in this study; Cu/Cd dosage response gradients have been conducted in maize (AbdElgawad et al. 2020; Atta et al., 202), but since no studies have been conducted in teosinte, we reasoned that it was important to apply the same treatment to both subspecies. We have now emphasized this rationale by adding the following in pg XX: “Whereas higher doses impair flowering or are lethal, lower Cu/Cd concentrations do not consistently show conventional phenotypic responses such as reduced plant growth (AbdElgawad et al. 2020; Atta et al., 2023)”.

We agree that the statement raised by the reviewer needed revision at the light of our results. We did revise the statement to accurately reflect our current evidence as follows: “Our results reveal a teosinte parviglumis environmental plasticity that is likely related to the function of HM response genes positively selected during the teosinte parviglumis to maize transition.”

(4) In maize, TEs are known to influence gene expression under abiotic stress, including for tb1 (PMID: 25569788). Since the author appears to make a causative conclusion between ZmHMA1, TB1, and HM stress, I would have liked to see a whole-transcriptome analysis and not a curation of two genes to determine whether other factors, such as TEs, can have that would lead to similar outcomes.

We agree that is definetely a possibility that we have not investigated at this stage. However, we added a pargraph to reflect this pertinent suggestion:

“Previous studies have demonstrated that transposable elements (TEs) contribute to activation of maize genes in response to abiotic stress, affecting up to 20% of the genes upregulated in response to abiotic stress, and as many as 33% of genes that are only expressed in response to stress (Makarevitch et al., 2015). It is therefore possible that the HM response of some specific genes that influenced maize emergence or domestication could be mediated by TEs influencing or driving their transcriptional regulation.”

(5) I would suggest that the authors carefully review the tables, figures, and the corresponding legends. For example :
a) Table 2 is called before Table 1, I would therefore suggest changing the numbering to reflect the paragraph order.

Thank you for your help, we did change the order of the Tables in the new version.

b) In Table 2, it is not clear whether the P value applies to the mean difference between WT and the mutant zmhma1, either in the presence or the absence of heavy metals. In addition, the authors need to use the P-value to estimate the differences between WT in the absence vs presence of HM, and WT in the absence of HM versus the mutant in the absence of HM (idem for presence).

We did address this issue in detail and added P-values and specific pairwise comparisons to that Table (now Table 1). Data are presented as mean ± standard deviation and were tested by a paired Student’s T-Test. When the effects were significant according to T-Test, the treatments were compared with the Welch two sample T-Test at P < 0.05.

c) Table 1 and Table 2: Indicate what type of statistical test was used and the number of plants used for each experiment (n). Also, I recommend the use of scientific notation for the P-values.

The statistical tests have now been indicated, scientific notation has been added to the P-values; the number of plants and biological replicates are indicated in the Methods section.

d) Lines 202 and 204: I assume Table 1 should be called instead of Table 2.

This error has been corrected.

e) General: In the text, when significance is highlighted along with measurements, the p-value needs to be added.

We have added the P-value along the measurement for all significant differences.

f) In the text, it is also mentioned that "the expression of ZMHMA1 was significantly increased in the presence of HMs (Figure 3c)". We are looking here at an RT-PCR, which is qualitative and without a robust quantitative comparison and statistics, I cannot conclude this assessment based on the presented evidence. No statistical measure is indicated here.

Panel 3c is not RT-PCR but a real-time qPCR, showing relative fold-change, normalized to actin, with a 3-technical triplicate per 3 biological replicates). We have added error bars (SD) and P-values represented by asterisks (calculated with Student's t statistic) to support significant differences (P<0.05 and P<0.01). ZmHMA1 expression was significantly increased in the presence of HMs only in teosinte; there was no significant difference in maize.

g) Figure 3 should at least have the gene name in the figure to quickly understand the figure panel. The key conserved domains should also be identified.

We agree and apologize for the omission. The gene names have been added adjacent to the structures.

h) Sentence at lines 459-460 lacks words and punctuation.

This unfortunate rror has also been corrected.

i) Figure S1, the reference Lemmon and Doebley, 2024 should be Lemmon and Doebley, 2014 to harmonize with the text.

The correct year is 2014. We have corrected this error.

Reviewer #2 (Recommendations for the authors):
(1) The narrative should be clearer, starting with a clearer hypothesis that is later sustained or not in the results, and then discussed in the idea and speculation section.

Thank you for the comment. We have clarified the hypothesis, it is included in the abstract and the last paragraph of the Introduction. We hope it is now clear that the evidence presented supports our hypothesis

(2) Focus more on traits that are relevant, for example, nodal and seminal roots.

We modified the text to emphasize three relevant traits. In the case of teosinte under HM stress, absence of tillering and increase in the number of female inflorescences. In the case of the zmha1 mutant under HM stress, differences in the number of nodal roots, and differences in height.

(3) RNA-seq in Cu/Cd stress could make the work much more useful and complete.

As previously mentioned, we have incorporated a large scale transcriptional analysis on the basis of six transcriptomes statistically validated (Table S5). Please see sections pgs. 11 to 13 for details.