The mouse embryonic fibroblast (MEF) cell lines: 10T1/2 (purchased from ATCC (Manassas, VA)), RHAMM^Δ163-overexpressing (LR21) cells, RHAMM^Δexon8-16 (KO), RHAMM^Δexon8-16 rescued with RHAMM^Δ163 (KOV4) and RHAMM^Δexon8-16 rescued with RHAMM^FL (KOV5) and the human breast cancer cell lines: MDA-MB-231 wild type and depleted of RHAMM using CRISPR/Cas9 were kind gifts from Dr. E. Turley, Western University, Canada and Dr. J. McCarthy, University of Minnesota. Because of different origin, the analysis was done by comparing 10T1/2 vs. LR21 and by comparing KO vs. KOV4 vs. KOV5. All the cell lines used in the study are telomerase positive. MDA-MB-231 cell line was used because telomerase activity is high in these human cells. The panel of murine cells were used to study the importance of N- and C- terminus of RHAMM and different isoforms of RHAMM. The cells were prepared and cultured to sub-confluency using Dulbecco’s modified Eagle’s medium (DMEM) (DMEM, 4.5 g/L Glucose; ThermoFisher, Cat # 21068028) containing 10% fetal bovine serum (FBS; Sigma Aldrich, cat #F0926), 4 μg/mL insulin, 8 μg/mL transferrin under the standard culture conditions of 37^oC in a humidified 5% CO₂ atmosphere as previously described (Tolg et al., 2003). The medium was changed every 2–3 days. The confluent cells were gently washed with sterile Phosphate Buffer Saline (PBS), pH 7.2–7.4, followed by harvested using TrypLE™ Express Enzyme (1X) (Gibco, Cat # 12563011).

The 10T1/2 fibroblasts were cultured in a complete medium for 24 h, then switched to a serum-starved medium (SSM) with 1% FBS. Cycloastragenol (CAG; Cat # SML1448, Sigma-Aldrich, United States) was added at 3 µM for 6 h, while control cells received DMEM in SSM. After 6 h, removing the medium, the cells were washed twice with cold PBS for qRT-PCR analysis. RHAMM^Δexon8-16 (KO), KOV4 and KOV5 were treated with PD98059 at 50 µM in SSM for 24 h before qRT-PCR analysis.

The TRAPeze Gel-based Telomerase Detection Kit (Cat #S7700, Millipore Sigma, United States) was used to perform the telomerase repeat amplification protocol (TRAP) assay according to the manufacturer’s instructions. Telomerase adds an AG sequence and telomeric repeats to the 3′end of substrate oligonucleotide (TS), followed by amplification using polymerase chain reaction (PCR) with unlabeled TS and reverse (RP) primers. This produces a ladder of products starting at 50 nucleotides in six-base increments. The PCR conditions were 94°C for 30 s, 59°C for 30 s, and 72°C for 1 min over 30 cycles. The products were analyzed on a 1% agarose gel, including controls for positive telomerase activity, PCR contamination, and a TSR-8 template as provided by the kit. Two independent experiments (n = 2) were conducted to validate the results.

The 10T1/2 fibroblasts were cultured in six-well plates with complete medium for 24 h, followed by serum starvation with 1% FBS for another 24 h. The conditioned medium was collected, and extracellular hyaluronan was quantified using the Hyaluronan Quantikine ELISA Kit (R&D, Cat # DHYAL0) according to the manufacturer’s instructions. In this assay, standards, controls, and samples were added to a microplate pre-coated with recombinant human (rh) aggrecan, allowing hyaluronan to bind to it. After washing to remove unbound substances, enzyme-linked rh aggrecan was added, followed by a substrate solution that developed color proportional to the bound hyaluronan. After stopping color development, the optical density was measured within 30 min at 450 nm. hyaluronan concentration was normalized to 1 μg of RNA extracted from cells, with data presented as means ± standard deviation (SD) from three replicates.

The significance of the hyaluronan binding and leucine zipper sequences was evaluated using a function-blocking RHAMM peptide, NP-110, in lung tissue from C57BL/6J mice. This tissue was sourced from a previous study (Wu et al., 2021) (kind gifts from Dr. E. Turley, Western University, Canada) in which the animal work was conducted by Stelic MC, Inc. in Tokyo, Japan, with institutional ethics approval and per the Guidelines for Proper Conduct of Animal Experiments (Science Council of Japan). In total, 36 six-week-old female C57BL/6J mice were randomly assigned to one of three groups: (i) the control group receiving 50 µL of 0.9% saline; (ii) the bleomycin (BLM) group receiving 50 µL of 1.5 mg/mL Bleomycin sulfate (Lot #15180, Nippon Kayaku, Japan) for 28 days; and (iii) BLM group receiving 3 mg/kg of NP-110 (sequence: ⁶⁴⁴KLKDENSQLKSEVSK) every fourth day (i.e., day 0, 4, 8, 12, 16, 20, 24) after BLM injections. Mice were subjected to sacrifice on day 28, and lungs were harvested and stored at −80 ± 5^oC for future use.

RNA isolation and qRT-PCR were performed according to established protocols (Tolg et al., 2017). Briefly, TRIZOL reagent (Ambion, United States) was used to isolate total RNA from the cells and lung tissue following the manufacturer’s instructions. Complementary DNA (cDNA) synthesis was conducted using iScript Reverse Transcription Supermix (BioRad, United States). Primers were designed with Primer three software (version 0.4.0) and synthesized by Life Technologies, United States (Supplementary Table S2). PCR reactions were set up with 1 μg of cDNA using the SsoAdvanced Universal SYBR Green Supermix kit (BioRad, United States). The PCR cycling conditions included an initial denaturation step at 95°C for 10 min, followed by 40 cycles of denaturation at 95°C for 30 s, annealing at 60°C for 1 min, and extension at 72°C for 1 min. All samples were analyzed in triplicate. Expression analysis and calculation of relative changes in gene expression were carried out using Stratagene Mx3000Pro software and MS Excel. Mouse gene expression was normalized to the Gapdh, Ldha, and Snrpd3 while human gene expression was normalized to GAPDH and HPRT1.

The Protein sequences of RHAMM from different animals, shown in Figure 2I were sourced from the NCBI Protein database (https://www.ncbi.nlm.nih.gov/protein/). NCBI-BLAST (https://blast.ncbi.nlm.nih.gov/Blast.cgi) checked the homology of the protein sequences of the ‘HATABD’ domain in RHAMM across the phylogenetic tree in the animal kingdom. Phylogenetic tree was drawn with iTOL (Letunic and Bork, 2024), and silhouettes are available on phylopic.org. We analyzed mRNA expression data of HMMR and hTERT using clinical datasets of ten different types of cancer retrieved from the TCGA database using cBioPortal (Cerami et al., 2012).

The difference between experimental groups of two-sample unequal variance was analyzed using Student’s t-test, with *p < 0.05, **p < 0.01, ***p < 0.001 being considered significant. The experiments were performed at least two to three times independently and represented as bar diagrams or dot plots using GraphPad Prism, ver. 10.3.1 (URL: https://app.graphpad.com).

The mRNA expression levels of RHAMM and TERT are positively correlated (p < 0.05) across ten types of human cancer and cancer cell lines, obtained from the TCGA Pan-Cancer Atlas database (Supplementary Table S1). To investigate this correlation, RHAMM was eliminated in the MDA-MB-231 cancer cell line using CRISPR/Cas9 (kind gift from E. Turley), leading to a significant 3-fold decrease (p < 0.01) in TERT mRNA expression (Figure 1B). A TRAP assay revealed that RHAMM-depleted cells had lower telomerase activity than those expressing RHAMM^Δ163 (Supplementary Figure S1A; original gel image shown in Supplementary Figure S3). Because RHAMM is expressed in multiple isoforms exhibiting different functions, further investigation revealed that RHAMM^Δ163, significantly increased TERT mRNA levels (p < 0.01; Figure 1C). However, TERT mRNA level remained unchanged in RHAMM^Δexon8-16 cells rescued with RHAMM^FL or depleted of RHAMM^FL and RHAMM^Δ163, suggesting that RHAMM^FL has no effect on TERT mRNA expression (Figure 1C). Furthermore, this result underscores that RHAMM C-terminus is crucial for TERT expression, but not the N-terminus. Similar patterns were observed for mRNA expression of dyskerin (Dkc1), which is critical for telomerase biogenesis (Cohen et al., 2007) (Supplementary Figure S1B). The TRAP assay confirmed that RHAMM^Δexon8-16 cells expressing RHAMM^Δ163 affected telomerase activity significantly compared to those with RHAMM^Δexon8-16 cells lacking RHAMM^Δ163 (Supplementary Figure S1A).

Telomere length is controlled through multiple mechanisms involving telomerase regulators and shelterin proteins (de Lange, 2005). PINX1 directly binds to TERT and inhibits telomerase catalytic activity endogenously (Zhou and Lu, 2001). Conversely, the NAD-dependent deacetylase sirtuin 1 (SIRT1) interacts with telomeric repeats and positively regulates telomere elongation (Palacios et al., 2010). SIRT1 directly interacts with TERT and regulates its nuclear localization and stability (Lee et al., 2024). In contrast, its abrogation results in increased telomere erosion. SIRT1 acts upstream of TPP1 and interacts with it (Lee et al., 2024) to regulate telomere function and cellular senescence. A reduction in TPP1 levels leads to telomere attrition and cellular senescence associated with SIRT1 (Ahmad et al., 2017). TPP1 forms complex with POT1 which is essential for recruiting telomerase to telomeres (Sekne et al., 2022). TPP1 does not bind to telomeric DNA but interacts with the telomerase essential N-terminal (TEN) domain, which is critical for the TTAGGG repeat addition processivity of telomerase and its recruitment to telomeres, while POT1 flexibly binds both telomeric DNA and telomerase. Together, TPP1 and POT1 enhance telomerase processivity by minimizing DNA dissociation during repeat synthesis, while TPP1-POT1 depletion impairs processive DNA synthesis. Hence, the TPP1-POT1 complex is vital for telomerase processivity (Sekne et al., 2022). This complex protects telomeres (Kibe et al., 2010).

The qRT-PCR analysis showed that RHAMM^Δ163 significantly reduced the negative regulator of TERT, Pinx1 mRNA levels while RHAMM^Δexon8-16 depleted of C-terminus RHAMM and RHAMM^Δexon8-16 rescued with RHAMM^FL increased Pinx1 expression (Figure 1D). Pinx1 expression was higher in C-terminus RHAMM-depleted cells compared to RHAMM^FL-expressing cells (p < 0.01). It could be possible that the N-terminus of RHAMM might enhance Pinx1 expression, while the C-terminus RHAMM represses it, which remains elusive and warrants further investigation. Thus, RHAMM^Δ163 may potentially activate TERT expression and lengthen telomeres, while RHAMM^FL is likely to inactivate TERT and shorten telomeres. Furthermore, RHAMM^Δ163 increased the mRNA levels of Sirt1 (positive regulator of TERT) while neither RHAMM^Δexon8-16 depleted of C-terminus RHAMM, nor RHAMM^Δexon8-16 rescued with RHAMM^FL had any effect (Figure 1E). This finding indicates that most likely, the N-terminus of RHAMM suppresses Sirt1 expression, while the C-terminus RHAMM elevates it. Eliminating RHAMM reduced TPP1 and POT1 mRNA expression in MDA-MB-231 (Figures 1F,I), indicating that RHAMM may regulate TPP1-POT1 complex. Further investigation revealed that RHAMM^Δ163 increased the mRNA levels of Tpp1 and Pot1a, while RHAMM^FL exerted no effect on these gene expression (Figure 1G). Taken together, these results show that RHAMM^Δ163 may promote telomerase activity by increasing SIRT1 and TPP1 expression and suppressing PINX1.

To further understand how RHAMM controls Tpp1 and Pot1a expression, RHAMM^Δexon8-16 (KO), RHAMM^Δexon8-16 rescued with RHAMM^Δ163 (KOV4) and RHAMM^Δexon8-16 rescued with RHAMM^FL (KOV5) were treated with PD98059 (50 µM), an MEK/ERK pathway inhibitor. As previously mentioned, RHAMM binds directly to ERK1 and indirectly to ERK2 and MEK through the sequence ⁷¹⁸LKQKIKHVVK⁷²⁷ at the C-terminus of RHAMM (mouse) (Tolg et al., 2010). This binding is vital for microtubule dynamics and hyaluronan- and CD44-mediated cell motility. The results show that PD98059 significantly decreased Tpp1 (Figure 1H) and Pot1a mRNA expression (Figure 1K) in KOV4 and KOV5 cells, while RHAMM^Δexon8-16 (KO) which lacks ERK1-binding domain (amino acids 718–727, mouse) promoted decreased Tpp1 and Pot1a levels. This result suggests that RHAMM^Δ163 may regulate TPP1 and POT1 expression through the ERK-mediated signaling pathway, possibly via the ERK1-binding domain, which needs further validation. Notably, RHAMM, CD44, and ERK1,2 form a complex that is required for activating ERK1,2 (Tolg et al., 2006). qRT-PCR analysis revealed a significant increase in Cd44 mRNA expression in RHAMM^Δexon8-16 rescued with RHAMM^Δ163 (KOV4) compared to RHAMM^Δexon8-16, with and without RHAMM^FL (Supplementary Figure S2A).

During mitosis, chromosome segregation relies on the organization of polymeric tubulin into bipolar spindles, with RHAMM decorating these spindles. As demonstrated in Figure 2A, the interaction of RHAMM with the cytoskeleton and centrosomes through TPX2 and AURKA is crucial for mitotic spindle assembly (Chen et al., 2014). Excessive AURKA upregulation disrupts normal mitosis, causing multinucleated cells and centrosome amplification. This is linked to TRF1, which stabilizes microtubule-kinetochore connections essential for chromosome segregation. AURKA phosphorylates TRF1, and excessive phosphorylation can lead to mitotic abnormalities (Ohishi et al., 2010). Notably, TRF1 is linked to TPP1 and POT1 via TIN2 and utilizes them to prevent ATR kinase during telomere replication and suppress sister telomere associations (Zimmermann et al., 2014).

RHAMM^Δ163 increased Trf1 levels significantly (Figure 2B). In contrast, RHAMM^Δexon8-16, with and without RHAMM^FL did not affect Trf1 expression (Figure 2B). The elimination of RHAMM decreased TRF1 mRNA expression in MDA-MB-231 cells indicating that RHAMM positively regulates TRF1 (Figure 2C). Ectopic expression of RHAMM^Δ163 elevated Tpx2 (Figure 2D) and Aurka mRNA levels (Figure 2E), provokes to hypothesize that RHAMM^Δ163 may regulate TRF1 transcription through TPX2-AURKA-mediated pathways, which needs further investigation. Notably, TRF1 interacts with BubR1, Nek2, Mad1, and Mad2 in the mitotic spindle checkpoint, playing a role in spindle formation (Muñoz et al., 2009; Prime and Markie, 2005). It is not clear if RHAMM and TRF1 act together in mitotic spindle regulation. However, evaluation of unpublished microarray analysis of 10T1/2 and LR21 cells (Tolg et al., 2012) indicates that RHAMM^Δ163 increases mRNA expression of Mad1l1, Mad2l1, and Nek2 (Table 1). This suggests that RHAMM^Δ163 affects TRF1-partners in mitotic spindle checkpoint (Figure 2E). Furthermore, RHAMM^Δ163 increases Rtel1 and Blm mRNA levels (Table 1). Notably TERF1 recruits these essential helicases, which resolve replication-associated issues and suppress the fragile-telomere phenotype. Additionally, RHAMM^Δ163 represses the mRNA level of telomerase-associated protein 1 (Tep1) (Table 1). Taken together, this pilot study suggests that RHAMM^Δ163 affects TERF1 expression, which may have paramount importance in telomere elongation and mitotic spindle regulation.

To investigate the importance of the ‘HATABD’ sequence of RHAMM on TERT mRNA expression, a mouse model with idiopathic pulmonary fibrosis (IPF) was chosen as IPF is characterized by TRF1 deletion or Tert deficiency, especially following a bleomycin (BLM) challenge (Povedano et al., 2015). HAS2 dysregulation in IPF (Li et al., 2016) and the effects of RHAMM antibodies and peptide mimetics in reducing macrophage recruitment and early fibrosis (Zaman et al., 2005) highlight the connection between RHAMM, hyaluronan, and IPF. In a prior study, the function-blocking RHAMM peptide, NP-110, which sterically disrupts RHAMM’s association with hyaluronan, TPX2, and AURKA (Figure 2F), was injected into BLM-treated mice, reducing fibrosis and increasing antifibrotic adipokines in skin (Wu et al., 2021). The current study analyzed the retained lung tissues from the previous study (a generous gift from Dr. E. Turley, Western University), revealing that NP-110 administration decreased the collagen 1a to collagen 3a ratio (Figure 2G) and elevated mTert mRNA expression almost three-fold (Figure 2H). These findings emphasize that blocking the interaction of RHAMM with hyaluronan, TPX2, tubulin, and AURKA via HATABD sequence may control IPF by regulating TERT expression.

The HATABD domain’s significance in regulating TERT expression by RHAMM prompted to screen the sequence across the animal kingdom in silico to understand if this sequence could be linked to longevity and telomere maintenance. The animals were selected based on their lifespan. The protein sequence homology of HATABD domain was examined in selected long-lived and short-lived species, using human RHAMM (HMMR) as a reference, with key amino acid changes highlighted in yellow (Figure 2I). The LKQKIKHVVK sequence in HABD1 is crucial for binding ERK1 (Tolg et al., 2010). Long-lived ectothermic reptiles showed two modifications (K636R and V644M), except crocodiles (demonstrating only V644M). These changes may influence RHAMM-ERK binding, although their connection to aging and telomere function is unclear. Stable HABD1 sequences were found in endothermic animals and long-lived naked mole rats. Additionally, the leucine zipper sequence (LKDENSQLKSEVSKL) contains three serine residues that were modified in long-lived ectothermic reptiles but not in short-lived species (Figure 2I). In-depth functional study is required to investigate the importance of these modifications on RHAMM interactions. Ectothermic tetrapods like crocodiles, turtles, and salamanders tend to age more slowly and have longer lifespans than birds and mammals (Reinke et al., 2022). This bioinformatic study prompts to hypothesize that amino acid substitutions in the ‘HATABD’ domain of RHAMM, under selection pressure, may relate to telomere maintenance and longevity, which warrants experimental validation.

To examine the impact of telomerase activity and TERT expression on RHAMM, 10T1/2 fibroblasts were treated with telomerase activator, cycloastragenol (CAG), a triterpenoid saponin from Astragalus membranaceus (Yang et al., 2023). This treatment significantly increased mRNA levels of Hmmr and Tert (p < 0.05) (Figures 3A,B) and enhanced mRNA expression of Tpx2 and Aurka (Figures 3C,D). It enhanced Has2 mRNA expression (Figure 3E), resulting in a nearly four-fold increase in hyaluronan synthesis in treated cells (600 ng/mL) compared to untreated cells (150 ng/mL, p < 0.05) (Figure 3fF. These findings suggest that telomerase activation and TERT upregulation may significantly enhance Hmmr expression and indicate a potential feedback loop with TERT, as noted for CD44 (Chung et al., 2013).