*4.6. Fruits*

Studies of *DUF642* gene expression and gene function during fruit development are very scarce. Of the Arabidopsis mutants studied, only *bdx-1* showed a fruit phenotype with shorter siliques and

lower seed production than wild-type plants. In the genus *Brassica*, *DGR2* is the only gene that has been detected in transcriptomes of siliques, specifically in the shells of the pods [44].

#### *4.7. Roots*

The expression of *DGR1*, *BDX*, *At5g11420*, *DGR2*, and *TEB* has been detected in primary roots of Arabidopsis. All these genes were expressed in the epidermal cells of the meristematic zone, except *DGR2*, which was expressed in the elongation zone. The *dgr2* mutant was the only one that presented a short root phenotype [17]. Overexpression of *AhDGR2* in Arabidopsis produced longer roots [16], which is understandable given that *DGR2* was expressed in the meristematic zone where the cells begin to expand and the rate of cell division is reduced.

The *bdx-1*, *At5g11420*, and *teb-1* mutants showed no alterations in the development of lateral roots. *BDX*, *At3g08030*, *At5g11420*, and *TEB* all had similar expression patterns throughout the development of lateral roots, which suggested there could be a functional redundancy that would explain the lack of altered phenotypes for these mutants.

Roots interact with microorganisms in the soil, and are particularly susceptible to pathogen attacks. Roots also are central in the perception of nutrient concentrations and other elements like heavy metals in the substrate where plants develop. The modification and restructuring of the cell wall in roots actively participate in the response to different types of stresses. Therefore, the participation of *DUF642* genes in the remodeling of the cell wall can be relevant in all these processes. Changes in the expression of *DUF642* genes in response to biotic and abiotic stresses have been described in various plant species (Table S2). However, very few functional studies are available related to the particular role of *DUF642* genes in response to stress.

#### **5. Biotic Factors**

*DGR2* is negatively regulated in Arabidopsis plants during infection by the bacterium *Ralstonia solanacerum* [45]. The expression of *At1g29980* and *At3g08030* is induced by infection with *Rhodococcus fascians*[46]. *At3g08030* also is highly expressed in response to infection by *Agrobacterium tumefaciens*[47].

The abundance of the DUF642 protein encoded by *At3g08030*, which has two isoforms with different isoelectric points, was reduced significantly in the cell wall proteome of Arabidopsis suspension cells treated with chitosan, an elicitor that mimics a fungal infection [48]. Infection with the *Penicillium* strain Pc4 in post-harvest grapes increased the abundance of two isoforms of the At5g11420 orthologous protein, probably because of post-translational modification [49]. In apoplastic proteomes of maize roots (*Zea mays*) infected with the symbiont species *Trichoderma virens*, a DUF642 protein was detected 5 days after inoculation. Modification and degradation of root cell walls are essential for colonization to take place [50]. Infection of Arabidopsis plants with the phytopathogenic fungus *Botrytis cinerea* reduced the expression of *At5g11420* and *At5g25460* [51].

The overexpression of *VqDUF642* in tomato plants reduced the susceptibility to *B. cinerea* infection in both mature and immature fruits. Overexpression of this gene modified the expression of some pathogen response genes such as *SIPR1*, *SIPR2*, *SIPR3*, and *SIPR4* whose expression increased at 48 h post-infection. In transgenic Thompson grape plants, overexpression of *VqDUF642* promoted increased resistance to *B. cinerea* and induced resistance to the fungus *Erysiphe necator,* which causes oidium of vines. The leaves of the overexpressing plants infected with *E. necator* showed a less severe infection than the wild-type at 48 h post-infection. In addition, the expression levels of the pathogen-response genes *VvPR1*, *VvPR2*, *VvPR3,* and *VvPR4* increased drastically at 48- and 96-h post-infection in transgenic seedlings compared with in the wild-type [15].

In Arabidopsis, transcriptomic induction during early inoculation with the knot-nematode *Meloidogyne incognita* revealed up-regulation of *At1g29980* [52]. *BDX* and *TEB* expression were also highly induced by *M. incognita* early inoculation. The cell wall localization of BDX and TEB in the epidermal cells of primary roots was induced by *M. incognita*. Early inoculation with *Nacobus aberrans*, a nematode that cannot infect Arabidopsis, did not alter the expression of these two genes [30].

A comparative analysis of available Arabidopsis flower transcriptomes showed that changes in the expression patterns of flower-specific defense genes were critical in pathogen resistance. According to this study, the expression of *BDX* and *DGR2* was positively regulated in petals compared with in senescent leaves, in stage 15 of the flower. At this stage, an increase in the expression of the biotic stress response genes also occurred [53].

### **6. Abiotic Factors**

The regulation of the DUF642 genes by transcription factors in response to aluminum stress has been reported. Aluminum stress has been studied extensively in aluminum-resistant plants. In particular, *Oryza sativa* (rice) is a cultivated crop plant that is resistant to aluminum, and the signaling pathway associated with its resistance has been studied widely. The Al resistance transcription factor 1 gene (*ART1*) is central to the aluminum response, and it regulates an increase in the expression of the *DUF642* gene *Os04g049490* (orthologous to *At5g11420*) [54]. *Os04g049490* was also regulated by the *SENSITIVE TO ALUMINUM RHIZOTOXICITY* gene (*STAR1*), which plays a fundamental role in aluminum resistance in rice roots [55]. In the *O. sativa indica* IR64 cultivar treated with aluminum, *Os04g41750* (orthologous to *At5g11420*) expression was up-regulated, whereas no change in its expression was detected in *O. sativa* cv. Azucena, which is more sensitive to aluminum [56]. Plants in the genus *Stylosanthes* have high tolerance to toxicity by the aluminum ion, and expression of the orthologous *DUF642* gene *Os04g0494900* increased in the roots of the Reyan 2 genotype in response to aluminum treatment [57]. A comparison between the transcriptomes of two *Citrus* species with different aluminum tolerances suggested that the process of HGs de-esterification in the cell wall of the root cells played an important role in resistance to aluminum. In *C. sinensis*, which is resistant to aluminum, four genes related to this process and a *DUF642* gene, an ortholog of *DGR2*, were up-regulated [58]. In NtSTOP-1-RNAi tobacco plants that showed a decrease in the expression of the *SENSITIVE TO PROTON RHIZOTOXICITY* gene (*STOP1*), the expression of *Nt6860* (orthologous to *At5g11420)* decreased, as has been described for *At5g11420* in the Arabidopsis *art-1* mutant [59].

In *Medicago sativa* plants exposed to cadmium, the amount of the protein orthologous to the Arabidopsis protein, DGR2, decreased [60]. The exposure of *Populus* × *canadensis* plants to high doses of zinc promoted the decrease in the expression of *POPTR\_0001s27110* (orthologous to *At3g08030*) [61].

Ultraviolet radiation (UV-B) generates an imbalance in the production of reactive oxygen species. Thus, it is used to study the effects of oxidative stress on plants. In the mutant *ggt1* of the protein GAMMA-GLUTAMYL TRANSFERASE ISOFORM 1(GGT1), involved in the redox balance, the proteins DGR2 and PECTIN METHYL ESTERASE3 (At3g14310) are very abundant in both the mutant and in the wt of Arabidopsis plants exposed to UV-B [62].

In the salinity-resistant plant *Manihot esculenta* Crantz (Cassava), *RknMes02\_00171* and *RknMes02\_00443*, which are orthologs of *BDX* and *At5g11420* respectively, were among the 40 most highly expressed genes during NaCl treatment [63]. *AhDGR2* was significantly induced in *Amaranthus hypochondriacus* (grain amaranth), which is also resistant to high doses of salt. However, when *AhDGR2* was overexpressed in Arabidopsis, the plants showed hypersensitivity to increasing NaCl concentrations, as shown by shorter root length, smaller and slightly chlorotic rosettes, as well as considerably reduced germination rates [16].

The expression levels of *At2g34510* and *At5g11420* increased in Arabidopsis plants subjected to drought stress, although their expression levels were not altered by treatment with abscisic acid [64]. The increase in *At2g34510* expression induced by drought was corroborated in a study that tested the stress response in natural populations of Arabidopsis [65]. In *Eucalyptus calmadulensis*, the expression of *Eucgr.C02812* (orthologous to *At5g11420*) was also found to decrease in plants exposed to drought [66].

#### **7. Conclusions**

The DUF642 proteins are highly conserved, which may be related to their specific interactions with cell wall polysaccharides and proteins in different cell types. The *At3g08030*-encoded protein interacts in vitro with cellulose, a polysaccharide that has an important role in plant development [9]. However, there are still no functional studies that confirm this interaction. BDX and At5g11420 interact in vitro with a PME [12]. Studies in different plant species suggest that their function could be related to the regulation of HGs modification throughout plant development. The functional studies carried out to date are consistent with this hypothesis. For instance, in almost all plant tissues where *DUF642* gene overexpression has been induced, there was an increase in the total PME activity. Furthermore, the increase in methyl-esterified HGs in the endosperm during embryo folding and in hypocotyl epidermal cells in the *DUF642* mutants is consistent with the corresponding phenotype. However, a direct interaction of DUF642 proteins with PMEs needs to be established using different experimental approaches. Several proteomic studies have indicated the presence of two isoforms of the same DUF642 protein, suggesting these proteins undergo post-translational modifications that may be related to differential interactions within the cell wall [29,60,61].

The different functions of *DUF642* genes during development can be determined by the specific spatial-temporal expression pattern for each gene. The differential response to hormones also may participate in this process. *BDX* is the only *DUF642* gene expressed in the endosperm during seed development. This specific expression may explain the misshapen-seeds phenotype in the *bdx-1* mutant. Similarly, the expression of *DGR2* in the root elongation zone may explain the short root phenotype exclusive of the *dgr2* mutant. It is possible that there is functional redundancy during embryo development as well as during the development and emergence of the lateral root, processes in which the same temporal-spatial pattern of at least two *DUF642* genes was detected in Arabidopsis.

*BDX*, *At5g11420*, *TEB*, and *At3g08030* are expressed in the meristematic zone of the primary root, but the *bdx* and *teb* mutants do not show an altered root phenotype. BDX was located intracellularly in the epidermal cells of the primary root and relocated to the cell wall in response to biotic and abiotic stresses. Roots interact with different organisms in the soil and are subjected to multiple perturbations in the environment, such as decreases in the availability of water and increases in toxic ions. These interactions can alter the expression of some DUF642 genes. In the response to fungal infection, the overexpression of *VqDUF642* increased resistance to the pathogen, although the expression other *DUF642* genes decreased in response to infection. These results suggest that the *DUF642* genes participate broadly in plant responses to environmental factors and not solely in the root developmental process.

The primary structure of *DUF642* proteins is highly conserved in different spermatophyte species. However, studies of their expression patterns in Arabidopsis showed that the spatial-temporal expression pattern for each gene was specific and consistent with the phenotypes of the mutant plants studied so far. The regulation of *DUF642* gene expression by hormones and environmental stimuli also was specific for each gene. Functional studies of the *DUF642* genes in different plant species are needed to determine the relevance of the DUF642 family in the evolution of terrestrial plants.

**Supplementary Materials:** Supplementary materials can be found at http://www.mdpi.com/1422-0067/20/13/ 3333/s1. Figure S1. *At3g08030* expression during *Arabidopsis thaliana* development using *pAt3g08030::ER-GFP* plants; Figure S2. *At5g11420* expression during *Arabidopsis thaliana* development using *pAt5g11420::ER-GFP* plants; Figure S3. *At4g32460*/*BDX, At5g25460*/*DGR2*, *At5g11420*, *At3g14310*/*PME3, At2g41800*/*TEB* and *At3g08030* expression during seed germination; Figure S4. Subcellular localization of BDX in *Arabidopsis thaliana* primary roots under salinity stress conditions using *pBDX::BDX-GFP* plants. Table S1. Transcriptomic evidence of *DUF642* genes expression during specific developmental stages and/or in response to diverse environmental stimuli. Table S2. Transcriptomic evidence of *DUF642* genes expression in response to diverse environmental stimuli.

**Author Contributions:** J.E.C.V., X.G.-M., and A.G.-d.; participated in the conceptualization and in the review and edition of the manuscript. J.E.C.V.; A.S.-I., E.Z.-S., A.H.-B., and E.Q.-R. carried out the molecular genetic studies. J.E.C.-V., X.G.-M., A.S.-I., E.Z.-S., and A.G.-d. participated in the original draft preparation. Funding was provided to A.G.-d. All authors read and approved the final manuscript.

**Funding:** This research was funded by Programa de Apoyo a Proyectos de Investigación e Innovación Tecnológica, Universidad Nacional Autónoma de México grant number IN203218- X.G.-M received a fellowship from Dirección General de Asuntos del Personal Académico, Universidad Nacional Autónoma de México.

**Acknowledgments:** We thank K. Jiménez for assistance on Confocal studies and M. Guemez for the artwork (Figure 1). We thank M. Biswas, from Edanz Group (www.edanzediting.com/ac) for editing a draft of this manuscript.

**Conflicts of Interest:** No conflict of interest is declared.
