Hidden Metabolic Trick Lets Myeloma Cells Steal Fuel to Survive a Key Cancer Drug

Bortezomib, a proteasome inhibitor that has anchored the treatment of multiple myeloma for nearly two decades, saves countless lives until it stops working. When patients relapse, their plasma cells somehow shrug off a drug designed to flood them with misfolded protein and trigger apoptotic collapse. A new study published in the Journal of Experimental & Clinical Cancer Research offers one of the most detailed mechanistic explanations yet for that failure, and it comes from an unexpected direction: a small molecule called putrescine, a transporter protein named SLC45A4, and a surprising supply line running from immune suppressor cells in the bone marrow directly into the cancer cells that are trying to kill their host.

The research team, led by Jun Xu, Peng Shu and Zhe Wang across institutions in Xi’an, Ningbo, Hefei and Anhui, began with a straightforward but technically demanding question. What actually differs, metabolically, between myeloma cells that respond to bortezomib and those that do not? They sorted CD138-positive malignant plasma cells from the bone marrow of five bortezomib-resistant and five bortezomib-sensitive patients at their institution and ran untargeted metabolomics by liquid chromatography with tandem mass spectrometry. Among the features that separated the two groups was a compound annotated as N,N’-diferuloylputrescine, a derivative in which both amino groups of putrescine are capped with feruloyl groups. That fingerprint pointed the investigators toward the polyamine family, a set of positively charged molecules long known to stabilize nucleic acids and support rapid growth.

Polyamines, which include putrescine, spermidine and spermine, are classic metabolic workhorses. Cells can either manufacture them from ornithine through the ornithine decarboxylase pathway or import them from their surroundings through an assortment of poorly characterized transporters. The team tested both routes. When they incubated sorted patient cells with deuterium-labeled spermine and spermidine and measured intracellular accumulation by mass spectrometry, resistant cells took up more polyamine from outside. Meanwhile, quantitative PCR of the genes governing polyamine biosynthesis, interconversion and transport showed that the synthetic machinery was not obviously cranked up in resistant samples. Blocking biosynthesis with difluoromethylornithine, a well-studied inhibitor of ornithine decarboxylase, did not meaningfully reshape the bortezomib dose-response relationship in the way an uptake-dependent mechanism would predict. The message was clear: resistant myeloma cells were not making more putrescine, they were importing it.

The transporter responsible turned out to be SLC45A4, a member of the solute carrier family 45 that had not previously held a starring role in myeloma biology. SLC45A4 messenger RNA was upregulated in bone marrow samples from bortezomib-resistant patients compared with sensitive ones, and in public datasets such as GSE4581 and GSE2658, higher SLC45A4 expression in pretreatment CD138-selected plasma cells correlated with worse overall survival in patients treated within the University of Arkansas Total Therapy programs. When the researchers forced SLC45A4 overexpression in four myeloma cell lines, XG-7, RPMI8226, U266 and MM1.S, the cells became markedly more tolerant of bortezomib across apoptosis assays, proliferation assays and dose-response curves. Conversely, knocking SLC45A4 down in KMS-11 cells with short hairpin RNA sensitized them to the drug, and rescuing the knockdown restored the resistant phenotype, a classic demonstration that the transporter is functionally required rather than merely correlated.

Isotope tracing then confirmed what the transporter was carrying. Cells overexpressing SLC45A4 accumulated significantly more deuterium-labeled putrescine from the medium, while their uptake of labeled spermine and spermidine and their intracellular pools of those larger polyamines were comparatively unaffected. The specificity matters because it distinguishes SLC45A4 as a putrescine importer rather than a general polyamine conduit. The team also noted that culturing cells in heat-treated fetal bovine serum, which would degrade labile extracellular factors, blunted the survival advantage, hinting that the relevant putrescine supply is found outside the cell, in the tumor’s microenvironment, rather than being generated internally.

Having established the import route, the investigators asked what the imported putrescine actually does inside the cell. This is where the study becomes genuinely novel. Using putrescine immunoprecipitation followed by mass spectrometry of the bound proteins, they found that the homeobox transcription factor HOXA1 was enriched in the putrescine-associated fraction, and that this association increased when SLC45A4 was overexpressed. Structural modeling with AlphaFold 3 placed putrescine in close proximity to residues F277 and Q278 of HOXA1. A Q278E point mutation, designed to disrupt that contact without globally misfolding the protein, selectively impaired putrescine binding while preserving HOXA1’s intrinsic transcriptional competence toward representative target genes such as POLR2I, HEBP1 and INCA1. Cellular thermal shift assays supported the idea that putrescine engagement changes HOXA1’s thermal stability inside cells, and this engagement persisted even when the crosslinking enzyme TGM2 was inhibited or replaced with a catalytically inactive mutant, arguing against an indirect TGM2-dependent explanation.

Functionally, putrescine-bound HOXA1 behaved as an amplifier of NF-kappaB signaling, a pathway with obvious relevance to myeloma survival and drug resistance. RNA sequencing of HOXA1-overexpressing cells showed enrichment of NF-kappaB-related genes, and dual-luciferase reporter assays demonstrated that HOXA1 boosted NF-kappaB-driven transcription in response to upstream activators such as MYD88 and TRAF6. Disrupting putrescine binding with the Q278E mutation, or depleting HOXA1 with guide RNA, dampened that activation. Western blotting showed corresponding changes in phosphorylated p65 and I-kappaB-alpha. In the clinical APEX dataset of relapsed myeloma patients treated with bortezomib or dexamethasone, higher HOXA1 expression tracked with poorer overall survival, and crossed perturbation experiments in which SLC45A4 and HOXA1 were manipulated individually and in combination confirmed that the two act cooperatively to drive putrescine accumulation and bortezomib resistance.

Perhaps the most striking finding concerns where the putrescine comes from. Single-cell RNA sequencing of bone marrow from a healthy donor, a bortezomib-sensitive patient and a bortezomib-resistant patient, supplemented by an independent public dataset, GSE189460, pointed to myeloid-derived suppressor cells, or MDSCs, as a rich source of polyamine uptake and synthesis activity within the resistant niche. Co-culture tracing experiments suggested that myeloma cells acquire putrescine from these immune suppressor cells, and exploratory cell-cell communication analysis hinted at intensified crosstalk in the resistant marrow. The authors are appropriately careful here: with only one individual per condition in the discovery cohort, those single-cell analyses are explicitly descriptive and hypothesis-generating, while the patient-level validation in the independent cohort treated each patient as a single biological replicate. Even with that caution, the picture that emerges is of a tumor ecosystem in which immunosuppressive myeloid cells feed malignant plasma cells a small molecule that hardens them against chemotherapy.

The translational payoff came from a pharmacological screen. Abiraterone acetate, a drug already approved for prostate cancer as a CYP17A1 inhibitor, preferentially suppressed myeloma cells overexpressing both SLC45A4 and HOXA1. Treatment with the compound reduced putrescine uptake of N15-labeled spermidine and spermine, destabilized SLC45A4 messenger RNA in an actinomycin D chase assay, and appeared to act independently of CYP17A1, since CYP17A1 knockdown did not reproduce the SLC45A4 downregulation. Combining abiraterone acetate with HXR9, a peptide inhibitor of HOX-PBX transcriptional complexes, attenuated the resistance-associated phenotype, and triple combinations with bortezomib produced the strongest effects in viability and proliferation assays. Genetic knockdown and rescue experiments confirmed that abiraterone acetate sensitivity genuinely depends on SLC45A4 and HOXA1 rather than on an off-target artifact.

The study carries the usual caveats of early translational work: the patient metabolomics cohorts are small, the single-cell discovery sample is n of one per condition, and abiraterone acetate’s repurposing would require formal clinical testing in myeloma before anyone could prescribe it for this purpose. But the conceptual contribution is substantial. Bortezomib resistance, the authors argue, is not purely a cell-autonomous property of the malignant clone; it is a niche-dependent metabolic negotiation, in which SLC45A4 imports putrescine supplied by MDSCs, putrescine engages HOXA1 at a defined molecular interface, and the resulting complex amplifies NF-kappaB signaling enough to blunt a proteasome inhibitor’s lethal pressure. If that axis holds up in larger cohorts, it suggests a genuinely new class of intervention for relapsed myeloma: not attacking the cancer cell’s genome or its proteasome, but cutting the metabolic supply lines that the bone marrow itself has laid down.

Subject of Research: SLC45A4-mediated putrescine uptake driving HOXA1 and NF-kappaB signaling in bortezomib-resistant multiple myeloma

Article Title: SLC45A4-mediated putrescine uptake functionally modulates HOXA1 and promotes bortezomib resistance in multiple myeloma

Article References: Xu, J., Shu, P., Zhang, L., Li, H., Xu, Z., Meng, B., Ma, J., Wang, W., Wang, W., Wu, X., & Wang, Z. (2026). SLC45A4-mediated putrescine uptake functionally modulates HOXA1 and promotes bortezomib resistance in multiple myeloma. Journal of Experimental & Clinical Cancer Research. https://doi.org/10.1186/s13046-026-03843-y

Image Credits: AI Generated

DOI: 10.1186/s13046-026-03843-y

Keywords: multiple myeloma, bortezomib resistance, SLC45A4, putrescine, HOXA1, NF-kappaB, polyamine metabolism, myeloid-derived suppressor cells, bone marrow niche, abiraterone acetate, drug resistance, cancer metabolism