The sponge R-RAS2-like protein, a single homolog of R-RAS subfamily members, is most closely related to the vertebrate R-RAS2 protein

The mammalian Ras-related subfamily encompasses R-RAS, R-RAS2, and M-RAS proteins. To elucidate the evolution of these proteins, we performed a phylogenetic analysis using representatives of the R-RAS subfamily from selected metazoans, as well as from their closest unicellular relatives. Among non-bilaterian animals (phyla Porifera and Cnidaria), a single homolog of this subfamily was identified, likely representing the ancestral type protein of all three members present in the human genome. Based on our bioinformatics analysis (Fig. 1), this single homolog is most closely related to the human R-RAS2 protein. Therefore, we named it R-RAS2-like protein.

Fig. 1

Evolutionary analysis of R-RAS subfamily proteins. A Phylogenetic relationships of R-RAS subfamily proteins in Metazoa and Protista. The representatives of R-RAS2-like and R-RAS2 proteins are highlighted in purple, with corresponding taxonomic groups indicated. The R-RAS and M-RAS proteins are shaded in grey. ML bootstrap values, based on 1000 bootstrapping replications, are represented as numbers associated with the branches (bootstrap values higher than 50% are displayed at the branching points). The scale bar denotes the number of substitutions per site. Additional file 2: Table S1 provides the accession numbers of the amino acid sequences used in our study. B Multiple sequence alignment of R-RAS2-like proteins from sponges and human R-RAS subfamily members. Conserved domains are indicated as follows: G motifs are shown in purple above the alignment, Switch regions in blue below the alignment, hypervariable region (HVR) in red below the alignment, and CaaX box in green above the alignment. C A schematic representation of human R-RAS2 protein with indicated conserved domains (above) and D the structure of r-ras2 and r-ras2-like genes from selected metazoans (below). Triangles mark the positions of introns, and the number within each triangle represents the intron phase. Black dashed lines connect introns that share the same positions and phases based on the alignment of amino acid sequences. The sequences of genes with corresponding intron positions were obtained from the NCBI's genomic database. E A table showing the number of R-RAS subfamily members among different species

Our phylogenetic analysis indicated that M-RAS likely diverged from this common ancestral R-RAS2-like protein within Bilateria, based on the presence of M-RAS homologs in lineages from annelids to mammals (Fig. 1A, E). The M-RAS proteins formed a distinct and well-supported clade (bootstrap value 100%) separate from other R-RAS subfamily members, indicating their independent evolution within the R-RAS subfamily. Our analysis showed that R-RAS and R-RAS2 diverged from a common R-RAS2-like protein much later in evolution, during the transition to Osteichthyes, indicating their potential functional specialization in Vertebrata (Fig. 1A, E). This divergence allowed the acquirement of unique properties of R-RAS and R-RAS2, enabling them to participate in specific cellular processes. Interestingly, birds exhibit a notable absence of the R-RAS protein, which is present in the genomes of most other vertebrate lineages (Fig. 1A, E). Avian genomes are characterized by a reduction in protein-coding genes and possess fewer members in some other gene families [36, 37]. Many of these genes play critical roles in inducing lethality in rodents, human genetic disorders, or exhibit tissue-specific biological functions [38]. In our phylogenetic analysis, R-RAS2-like proteins from sponges formed a clade with homologs from protists closely related to animals. Additionally, these proteins formed a sister group with homologs from other non-bilaterian animals. Representatives of R-RAS2-like proteins from bilaterians were placed in a separate clade. A divergence within vertebrates resulted in the appearance of two already mentioned paralogs, R-RAS and R-RAS2. The branches representing these paralogs showed high bootstrap values (99% and 100%, respectively), which support the inferred relationships in the phylogenetic tree.

The R-RAS2-like protein identified in the sponge E. subterraneus comprises 191 amino acids. As sponges have only one R-RAS subfamily member, we conducted a comparative analysis to determine its similarity to the three corresponding members in the human genome. Sponge R-RAS2-like protein shares the highest homology (78.4%) with human R-RAS2, followed by 71.6% homology with M-RAS and 64.2% homology with R-RAS (Additional file 3). To further investigate the evolutionary conservation of regions involved in the GTP binding and hydrolysis, which are common among all RAS superfamily members, we compared homologs from sponges and humans. We found that G-motifs are highly conserved between sponges and human R-RAS subfamily members, with minor differences observed in the G1 and G4 motifs of the human M-RAS protein (Fig. 1B). This observation is consistent with our phylogenetic analysis, as M-RAS represents the most evolutionary distant member of the R-RAS subfamily. In addition, we found that the SwitchI region is highly conserved (100% identity across all analyzed sequences), while the SwitchII region is also conserved, although showing lower sequence identity. Switch regions are highly dynamic and change their conformation upon GTP binding and hydrolysis. Of particular note are the conserved SwitchI threonine (T46 in HsaRRAS2, corresponding to T35 in canonical Ras proteins) and SwitchII glycine (G71 in HsaRRAS2, corresponding to G60 in canonical Ras proteins) which form hydrogen bonds with the γ-phosphate and hold SwitchI and SwitchII regions in the active conformation, respectively. Upon GTP hydrolysis, the γ-phosphate is released and both Switch regions return to the flexible conformation in the GDP-bound state [5, 39]. Significant differences in protein length were observed among members of the R-RAS subfamily, particularly the elongated N-terminal region of human R-RAS and the hypervariable region (HVR) at the C-termini of the analyzed proteins. Notably, within the HVR, both human R-RAS and R-RAS2 exhibit a conserved proline-rich motif, commonly referred to as the R-RAS box, which is absent in human M-RAS and the R-RAS2-like homologs in sponges. Following this proline-rich motif is the CaaX box, a sequence present in all representatives of the R-RAS subfamily, known to be crucial for their localization to the plasma membrane [6]. We further aligned the R-RAS2 and R-RAS2-like homologs from metazoans and their closest unicellular relatives, and our results confirmed significant conservation of the five G-motifs and two switch regions that are crucial for GTPase activity (Additional file 2: Fig. S1). Moreover, we observed that the proline-rich motif is present only in vertebrate R-RAS2 representatives and absent in R-RAS2-like homologs from lower metazoans. This indicates that the proline-rich region likely emerged during the divergence of R-RAS and R-RAS2 paralogs from a common ancestral R-RAS2-like protein. We also identified that the CaaX box, which plays an important role in protein localization, is conserved among all R-RAS2 and R-RAS2-like proteins, emphasizing its functional significance. A heatmap displaying multiple sequence alignments revealed a high overall protein sequence homology among R-RAS2 and R-RAS2-like proteins (Additional file 2: Fig. S2). The identity/similarity scores exceeded 50%, indicating evolutionary conservation and the significant cellular role of this protein. As expected, the highest homology was observed among vertebrate R-RAS2 proteins, ranging from 75 to 100% (Additional file 4).

We conducted a comprehensive structural analysis of r-ras2 genes from selected vertebrates, along with r-ras2-like homologs from lower metazoan lineages (Fig. 1C, D). Considering that r-ras and m-ras genes emerged through divergence from a common r-ras2-like gene during animal evolution, we included human r-ras and m-ras genes in our analysis. We compared the intron–exon composition to determine whether the conservation of gene structure aligns with the observed protein conservation. Our results revealed that the human r-ras2 gene possesses all five introns that were originally present in the common ancestral r-ras2-like gene found in non-bilaterian animals. Notably, three of these introns are shared between sponges and humans, demonstrating a conserved intron composition across these phylogenetically distant species. We also observed identical positions and phases of all introns within human r-ras and r-ras2 genes, suggesting a close evolutionary relationship and relatively recent divergence. Interestingly, the human m-ras gene shares two introns with other human paralogs, while the remaining two introns are exclusively shared with r-ras2-like genes from non-bilaterian animals and are absent in r-ras2-like genes from bilaterians (Fig. 1D). These findings support our previous results and reinforce the hypothesis that m-ras diverged from the r-ras2-like gene during the transition from non-bilaterians to bilaterians. Our results contribute to a deeper understanding of the evolutionary dynamics and the conservation of gene structure within the r-ras gene subfamily.

The sponge R-RAS2-like and human R-RAS2 have similar biochemical properties

To analyze the biochemical properties of the EsuRRAS2-like protein and to compare it with human R-RAS2, we produced and purified both proteins (Fig. 2A). The GTPase activity of the sponge homolog was measured using the luminescence-based GTP hydrolysis assay. Initially, we titrated the HsaRRAS2 protein in the presence of 1 µM GTP to determine the optimal enzyme concentration for the reaction (Fig. 2A). Next, we analyzed the intrinsic GTPase activity of EsuRRAS2-like and HsaRRAS2 at concentrations of 6.25 µM. Using a luminescence-based GTP hydrolysis assay, we observed a significant reduction in luminescence signal for both EsuRRAS2-like and HsaRRAS2, indicating GTP hydrolysis (Fig. 2B). This demonstrated that ancient EsuRRAS2-like already possesses intrinsic GTPase activity, similar to its human homolog, and indicates a conserved role of GTPase activity of R-RAS2 in regulating similar signalling pathways and cellular processes in sponge and humans.

Fig. 2

Biochemical properties of EsuRRAS2-like and HsaRRAS2 proteins. A Titration of EsuRRAS2-like and HsaRRAS2 proteins in GTPase/GAP Buffer with a molar concentration of 1 µM GTP. Successful isolation and purification of EsuRRAS2-like and HsaRRAS2 proteins confirmed by SDS-PAGE gel. Concentration of 6.25 µM (791.875 ng) is marked with rectangle. B Both the sponge and human R-RAS2 homologs exhibited intrinsic GTPase activity at a concentration of 6.25 µM. Luminescence was measured after a two-hour incubation. The control sample contained only the GTP/GAP buffer. Standard deviations are indicated as mean ± SD, n = 3. RLU (relative luminescence unit). ***p < 0.005. C RNA binding activity of sponge R-RAS2-like and human R-RAS2. Sponge and human proteins (5 µg) were preincubated with increasing concentrations of free poly(U), followed by incubation with 50% poly(U) agarose beads. The RNA binding protein DRG1 served as a positive control, while BSA served as a negative control. After incubation, proteins were analyzed by SDS-PAGE and stained with Coomassie Brilliant Blue. Abbreviations: I-input, B-beads

To evaluate the RNA-binding ability of sponge and human R-RAS2 proteins, we used polyuridylic acid (poly(U)) agarose beads, as described in previous studies [32, 40]. Our results show that both EsuRRAS2-like and HsaRRAS2 proteins exhibited binding affinity towards the poly(U) agarose beads, whereas BSA, which served as the negative control, did not. To further investigate whether RNA binding is specific, we introduced free poly(U) as a competitor during the binding assay. We observed a dose-dependent reduction in binding when R-RAS2 proteins were pre-incubated with increasing concentrations of free poly(U) before the addition of poly(U) beads (Fig. 2C). These results indicate that the interaction between R-RAS2 proteins and RNA is specific.

The sponge R-RAS2-like and human R-RAS2 have similar but not identical localizations

To determine the R-RAS2 protein localization in human tumor cells, we co-transfected MCF-7 and HeLa cells with EsuRRAS2-like fluorescently labelled with CHERRY, and HsaRRAS2 fluorescently labelled with GFP. We observed that the exogenous sponge and human homologs localize in the cytosol of MCF-7 and HeLa cells and not in the nucleus. The punctuate staining patterns indicate the association of EsuRRAS2-like and HsaRRAS2 with the plasma membrane and other intracellular membranes, likely the endocytic vesicles. Based on the staining morphology, the sponge homolog is more localized in intracellular membranes, whereas the human homolog is more localized in the plasma membrane (Fig. 3A and Additional file 2: Fig. S3). Therefore, we confirmed partial colocalization of EsuRRAS2-like and HsaRRAS2 in the membranes of MCF-7 and HeLa cells. The localization of EsuRRAS2-like mainly at endosomal membranes rather than at the plasma membrane may be due to two differences between these proteins. The first is the lack of a cysteine residue crucial for palmitoylation, while the second is a shorter HVR in the sponge EsuRRAS2-like protein (Fig. 3B), as already shown for human R-RAS [6].

Fig. 3

Sponge and human R-RAS2 homologs have similar but not identical localization in membranes of MCF-7 cells. A Colocalization (yellow) of human R-RAS2 with sponge homolog EsuRRAS2L in the membranes of human breast cancer cells MCF-7. Human R-RAS2 was fluorescently labelled with GFP, and sponge EsuRRAS2L was fluorescently labelled with CHERRY (red). Hoechst was used to stain nuclei. The experiments were repeated three times in biological duplicates. Cells were analyzed by confocal microscopy. B The C-terminal hypervariable region of R-RAS subfamily members. The proline-rich region, present in HsaRRAS and HsaRRAS2 proteins, is shown in yellow, while the cysteine residue that undergoes palmitoylation is shown in orange. Lysine residues conserved between sponge R-RAS2-like and human R-RAS2 are shown in green. The CaaX box conserved between all analyzed sequences is highlighted in pink. Esu-sponge Eunapius subterraneus, Hsa-human. Scale bar—10 μm

We further analyzed the specific membranes in which EsuRRAS2-like and HsaRRAS2 proteins are localized. For that purpose, we co-stained MCF-7 cells transfected with GFP-labelled sponge or human R-RAS2 homologs with markers for endocytic vesicles (Fig. 4). We observed partial colocalization of EsuRRAS2-like with early endosomes (EEA1), which was not observed for HsaRRAS2 (p*** < 0.001, Fig. 4A, E). Our results demonstrated that both EsuRRAS2-like and HsaRRAS2 proteins exhibited the highest colocalization with recycling (TfR) and late endosomes (Rab7) (Fig. 4B, C, E) and no colocalization with lysosomes (LAMP1, Fig. 4D, E). The localization within recycling endosomes was confirmed using additional marker, RAB11 (Additional file 2: Fig. S4A, B). These results provide valuable insight into the precise subcellular localization of EsuRRAS2-like and HsaRRAS2 within the endosomal pathway.

Fig. 4

Both sponge and human R-RAS2 are localized within recycling and late endosomes. Colocalization (yellow) of sponge EsuRRAS2L or human HsaRRAS2 fluorescently labelled with GFP (green) with markers (red) for A early endosomes (EEA1), B recycling endosomes (TfR), C late endosomes (Rab7), and D lysosomes (LAMP1) in human breast cancer cells MCF-7. Hoechst was used to stain nuclei. E Quantification of colocalization between EsuRRAS2L or HsaRRAS2 with endosomal markers was done using ImageJ Coloc2 plugin and shown as Pearson’s correlation coefficient. ***p < 0.001, n = 30 cells per group from three different experiments. Cells were analyzed by confocal microscopy. Esu-sponge Eunapius subterraneus, Hsa-human. Scale bar—10 μm

In order to distinguish whether these proteins participate in early or late endocytosis, we further wanted to analyze the localization of both EsuRRAS2-like and HsaRRAS2 within vesicles in MCF-7 cells. For this purpose, we co-transfected CHERRY-tagged EsuRRAS2-like or HsaRRAS2, along with GFP-tagged Rab5 (an early endosome marker, Fig. 5A) or Rab7 (a late endosome marker, Fig. 5B). Although we observed only partial colocalization within early endosomes, EsuRRAS2L shows statistically more colocalization with Rab5 than HsaRRAS2 (p*** < 0.001, Fig. 5A, C), our results confirmed the localization of EsuRRAS2-like and HsaRRAS2 within late endosomes (Fig. 5B, C). These results are in accordance with endosomal transport of proteins localized in both plasma and endosomal membranes [41].

Fig. 5

Both sponge and human R-RAS2 are localized within overexpressed early and late endosomal marker proteins. Colocalization (yellow) of cotransfected sponge EsuRRAS2L or human HsaRRAS2 fluorescently labelled with CHERRY (red) together with markers (green) for A early endosomes (Rab5), and B late endosomes (Rab7) in human breast cancer MCF-7 cells. Hoechst was used to stain nuclei. C Quantification of colocalization between EsuRRAS2L or HsaRRAS2 with endosomal markers was done using ImageJ Coloc2 plugin and shown as Pearson’s correlation coefficient. ***p < 0.001, n = 30 cells per group from three different experiments. Cells were analyzed by confocal microscopy. Esu-sponge Eunapius subterraneus, Hsa-human. Scale bar—10 μm

To investigate the potential impact of EsuRRAS2-like and HsaRRAS2 overexpression on the expression levels of proteins associated with early, recycling, and late endosomes, and lysosomes (EEA1, TfR, Rab 11, Rab7, and LAMP1), we transfected MCF-7 cells with FLAG-tagged EsuRRAS2-like or HsaRRAS2 constructs and analyzed proteins of interest by Western blot analysis. We found that overexpression of HsaRRAS2 significantly reduced the levels of TfR (a marker for recycling endosomes, p < 0.001, Fig. 6A, C) compared to cells transfected with an empty vector. The overexpression of EsuRRAS2-like had a similar effect of reducing the levels of TfR (p = 0.0158) (Fig. 6B, D). However, neither HsaRRAS2 nor EsuRRAS2L change the expression levels of Rab11, another marker of recycling endosomes (Additional file 2: Fig. S4C, D, E, F). Furthermore, the overexpression of HsaRRAS2 caused increased protein levels of EEA1 (p = 0.0011) and decreased levels of LAMP1 (p = 0.0032) (Fig. 6A, C) while the sponge homolog EsuRRAS2-like did not (Fig. 6B, D). Therefore, human homolog impacts TfR level more than the sponge homolog, and the sponge homolog does not affect EEA1 and LAMP1 levels. Although our study shows that both exogenous HsaRRAS2 and EsuRRAS2-like influence the dynamics and turnover of endosomal vesicles in MCF-7 cells, further studies are necessary to clarify these observations.

Fig. 6

The expression of human and sponge R-RAS2 alters the expression of endosomal vesicles markers. Levels of overexpressed A human or B sponge R-RAS2 homologs labelled with FLAG, and endogenous levels of markers for endosomal vesicles: early endosomes (EEA1), lysosomes (LAMP1), recycling endosomes (TfR) and late endosomes (Rab7) were analyzed by Western blot and detected with specific primary antibodies. C and D Quantification for each protein levels was shown as ratio to empty-vector, *p < 0.05, **p < 0.01, ***p < 0.001. Amido Black was used as a loading control. Cropped blots are displayed. Esu-sponge Eunapius subterraneus, Hsa-human

Main biological functions of the sponge R-RAS2-like and human R-RAS2 are conserved

Next, we investigated the biological effects of EsuRRAS2-like on cancer cells. For this purpose, we used MDA-MB-231 cells, the most commonly used triple-negative breast cancer cell line model, which carry mutations in Ras effector pathway (BRAF and NF1 mutations). First, we analyzed the proliferation of cells transfected with FLAG-tagged EsuRRAS2-like and HsaRRAS2 constructs. A significant increase in cell proliferation was observed for both EsuRRAS2-like and HsaRRAS2 (p < 0.001) compared to an empty-vector (Fig. 7A). Next, we examined the impact of EsuRRAS2-like or HsaRRAS2 transfection on cell survival and colony formation. Similar to the cell proliferation results, both EsuRRAS2-like and HsaRRAS2 overexpression increased the number of colonies formed (p < 0.001, respectively) (Fig. 7B). To assess the role of EsuRRAS2-like and HsaRRAS2 in cell migration, we employed wound healing and Boyden chamber assays. MDA-MB-231 cells transfected with both EsuRRAS2-like and HsaRRAS2 exhibited enhanced cell migration and faster wound closure compared to control (p < 0.001) (Fig. 7C, D). These results confirm the conserved function of human R-RAS2 and its sponge homolog in tumor-related processes, indicating their potential oncogenic role. This reiterates the functional significance of R-RAS2 in cancer cells and highlight the conservation of its biological functions across animals.

Fig. 7

Both sponge and human R-RAS2 homologs display oncogenic properties in MDA-MB-231 cells. A cell proliferation, B number of colonies formed, C wound healing, and D cell migration was quantified using the ImageJ software (National Institutes of Health, USA). The statistical significance of the tests was set at *p < 0.05, **p < 0.01, ***p < 0.001. The experiments were repeated three times in biological duplicates. Esu-sponge Eunapius subterraneus, Hsa-human

The sponge R-RAS2-like and human R-RAS2 regulate focal adhesions

It has recently been shown that endogenous R-RAS2 (upon R-RAS2 knock-out and knock-in of fluorescently labelled R-RAS2), unlike canonical Ras proteins, specifically localizes in focal adhesions [12]. However, in our experiments we did not observe that the sponge or the human homolog localizes in focal adhesions. In contrast, our data showed the association of EsuRRAS2-like and HsaRRAS2 with the plasma membrane and endocytic vesicles. More specifically, both EsuRRAS2-like and HsaRRAS2 proteins showed the highest colocalization with recycling (TfR) and late endosomes (Rab7) (Figs. 4B, C, E and 5B, C). To demonstrate the ability of transfected HsaRRAS2 or EsuRRAS2-like to regulate focal adhesions, we performed biochemical isolation of focal adhesions in MDA-MB-231 cells transfected with HsaRRAS2-FLAG or EsuRRAS2-like-FLAG constructs. The method is based on the use of the crosslinker DTBP, which diffuses into cells and crosslinks preferentially adhesion proteins. Cells are then lysed, washed with high pressure tap water to remove cell bodies, and focal adhesions are collected by scraping from Petri dishes [33]. All experiments were performed without prior coating of the growing surface with extracellular matrix proteins, as described in [34]. The samples were analyzed by Western blot using antibodies specific for two classic focal adhesion proteins, talin1 [42] and vinculin [43] as well as anti-FLAG antibodies. The overexpression of HsaRRAS2 in MDA-MB-231 cells significantly increased the level of talin1 and vinculin within focal adhesion isolates compared to cells transfected with the empty vector. A similar effect, although not as strong, was also observed after expression of EsuRRAS2-like (Fig. 8A). This result is consistent with the effect of transfected EsuRRAS2-like or HsaRRAS2 on cell proliferation, migration and colony formation (Fig. 7). MDA-MB-231 cells preferentially use integrin αVβ5 for adhesion [44] and for this reason we analyzed integrin subunit β5 expression within focal adhesion isolates. Surprisingly, the expression levels of integrin β5 did not change (Fig. 8A). Results of Western blot analysis obtained from pooled data from three independent experiments confirmed these observations (Fig. 8B). Due to the nature of the experiment involving transfection, lysis, cross-linking, washing and Western blot, standard deviations are large which affects statistical analysis. We were unable to demonstrate EsuRRAS2-like-FLAG or HsaRRAS2-FLAG expression in focal adhesion isolates, using anti-FLAG antibodies, indicating that EsuRRAS2-like and HsaRRAS2 do not localise within but regulate the formation of focal adhesions, however, not those formed by integrin heterodimers αVβ5.

Fig. 8

Both sponge and human R-RAS2 homologs regulate focal adhesions in MDA-MB-231 cells. A Twenty-four hours upon transfection with either empty vector of vector containing EsuRRAS2-like-FLAG or HsaRRAS2-FLAG construct, focal adhesions were isolated and WB analysis of talin1, vinculin and integrin β5 was performed. Numbers below blots represent the relative expression of proteins compared to control (empty-vector) normalized against amidoblack staining of focal adhesion isolates. Densitometry was done using the ImageJ software (National Institutes of Health, USA). B Quantification of Western blot data presented in (A) together with two independently performed biological replicas. Histogram data are plotted as mean ± SD (n = 3) relative to expression in cells transfected with an empty-vector that was set as 1 (indicated by a dotted line). Data were analyzed by unpaired Student’s t-test. *p < 0.05. C Forty-eight hours after transfection with either an empty-vector or a vector containing EsuRRAS2-like-GFP or HsaRRAS2-GFP construct, cells were fixed with paraformaldehyde and incubated with Alexa-Flour 488 conjugated phalloidin for F-actin visualization, anti-integrin β5 antibody followed by Alexa-Fluor 647-conjugated antibody and IRM images were taken. Analysis was performed using TCS SP8 Leica. Scale bar—10 μm. Esu-sponge Eunapius subterraneus, Hsa-human

To further investigate whether transfected R-RAS2 homologs regulate focal adhesions, we transfected MDA-MB-231 cells with HsaRRAS2 or EsuRRAS2-like fluorescently labelled with GFP and visualized the F-actin cytoskeleton and integrin αVβ5 focal adhesions. At the same time, IRM was used to additionally visualize focal adhesions (Fig. 8C). Again, we did not find the fluorescently labelled R-RAS2 homologs in focal adhesions. However, due to the expression of HsaRRAS2 or EsuRRAS2-like, we observed slightly altered appearance of actin stress fibers compared to cells transfected with a control plasmid. In our hands, the quantification of integrin αVβ5 focal adhesions or actin stress fibers in MDA-MB-231 cells was not possible. However, a melanoma cell line MDA-MB-435S also preferentially uses integrin αVβ5 for adhesion [34] and contains a very large number of focal adhesions with highly pronounced stress fibers. Therefore, to further analyze whether any of R-RAS2 homologs localize in and/or regulate focal adhesions, we visualized transfected HsaRRAS2 or EsuRRAS2-like fluorescently labelled with GFP together with actin stress fibers. The IRM and also integrin β5 were used to visualize focal adhesions. Similarly, as we observed in MDA-MB-231 cells, in MDA-MB-435S cells we could not find the fluorescently labelled R-RAS2 homologs in focal adhesions, but similarly we observed slight changes in organisation of F-actin (Additional file 2: Fig. S5A). However, quantification of integrin αVβ5 focal adhesions and actin stress fibers in MDA-MB-435S cells transfected with HsaRRAS2 or EsuRRAS2L showed that the number and size of integrin αVβ5 focal adhesions did not change, nor did the total amount of actin stress fibers (Additional file 2: Fig. S5B). Therefore, the sponge and human R-RAS2 homologs, HsaRRAS2 and EsuRRAS2-like, do not localise within focal adhesions but play a role in their regulation, but not those formed by integrin heterodimer αVβ5. It remains to be investigated which focal adhesions are altered and the sequence of events that leads to increased migration after overexpression of either HsaRRAS2 or EsuRRAS2L.