Single-cell transcriptional profile of CD34+ hematopoietic progenitor cells from del(5q) myelodysplastic syndromes and … – Nature.com

Single-cell RNA-sequencing of hematopoietic progenitor cells of del(5q) MDS patients

To identify the transcriptional alterations characterizing hematopoietic progenitors harboring del(5q), we initially performed scRNA-seq of CD34+ cells of four newly diagnosed patients with del(5q) MDS (Patient_1-4), and three age-matched healthy donors (Healthy_1-3) using the 10X Genomics technology (Fig.1A) (gating strategy can be found in Supplementary Fig.1). The clinical and genomic characteristics of the MDS patients and healthy donors are shown in Supplementary Table1. The percentage of cells with del(5q) based on the cytogenetic analysis varied between 35 to 90%. In all cases, the common deleted region encompassed bands 5q(13-33) (genes shown in Supplementary Data1).

A CD34+ cells were obtained from bone marrow aspirates of newly diagnosed del(5q) MDS patients, healthy donors and patients treated with Lenalidomide, and were subjected to single-cell RNA sequencing and analysis. PCR partial cytogenetic responder, CCR complete cytogenetic responder, NR non-responder. Part of this figure was created with BioRender.com. B Uniform Manifold Approximation and Projection (UMAP) of 42,494 cells representing the expected 14 hematopoietic progenitors: HSC hematopoietic stem cells, LMPP lymphoid-primed multipotent progenitors, GMP granulocyte-monocyte progenitors; granulocyte progenitors; monocyte progenitors; dendritic cell progenitors; CLP common lymphoid progenitors; B-cell progenitors; T-cell progenitors; MEP megakaryocyte-erythroid progenitors; MK_Prog megakaryocyte progenitors; early erythroid progenitors; late erythroid progenitors; basophil progenitors. C Per patient UMAP showing the identity of the cells projected from the integrated space. D Dotplot showing the percentage and value of the normalized expression of the canonical marker genes used to assign the cell identity to each cluster. E Barplot representing the contribution of cells from each patient to the different cell types. F Barplot representing the number of cells assigned to each cell type for the studied patients. G Barplot representing the percentage of cells assigned to each cell type for del(5q) MDS patients and healthy samples. N=7 biologically independent samples were used (n=3 healthy donors and 4 del(5q) MDS patients). Data are presented as mean values +/SD. Two-sided Wilcoxon signed-rank test was used to calculate statistical significance. Exact p-values for the differential abundance of each hematopoietic progenitor between the del(5q) MDS and the healthy condition were the following: HSC: p=0.63; LMPP: p=0.23; GMP: p=0.06; Granulocyte: p=0.23; Monocytes: p=0.06; DendriticCell: p=0.63; CLP: p=1; pro-B: p=0.23; T: p=0.04; MEP: p=0.11; MK_Prog: p=0.63; EarlyErythroid: p=1; LateErythroid: p=0.23; Basophil: p=0.4.

A total of 55,119 and 45,311 cells from patients and healthy donors, respectively, were profiled and integrated. After applying quality filters, 46,772 and 43,442 cells were eventually included in the downstream analysis. Data was integrated, clustered and manually annotated (Fig.1B, C, Supplementary Fig.2A, B) based on curated markers (Fig.1D, Supplementary Fig.2C), obtaining 14 and 13 clusters (patients and donors, respectively) representing all the expected hematopoietic progenitor subtypes. Contribution of every MDS patient and donor to the composition of all the clusters was identified (Fig.1E, Supplementary Fig.2D), and each individual showed different proportions of hematopoietic progenitors (Fig.1F, Supplementary Fig.2E). Although there were some differences in the percentage of hematopoietic progenitors between MDS patients and healthy donors (e.g., HSC), these differences were not statistically significant which might be related to the high variability in cell composition across samples (Fig.1G).

Identifying single-arm copy number variations (CNVs) at the single-cell level presents challenges due to potential compensatory mechanisms of alleles, as well as to the sparse and noisy nature of single-cell data. In this study, we employed two different and complementary approaches: CopyKat23 (Fig.2A), which relies on gene expression, and CaSpER24 (Fig.2B), which relies on allele frequencies (see Methods). This combined strategy aimed to enhance the sensitivity and accuracy of identifying cells harboring 5q deletion. To avoid false positive detection, we only classified the cells as harboring the del(5q) if the same cell was characterized as such by the two different algorithms (Fig.2C). To validate this classification, we analyzed the expression pattern of genes encoded in the deleted region in individual cells. Due to the sparsity of scRNA-seq data, we were only able to detect six genes as highly variable, CD74, RPS14, BTF3, COX7C, HINT1 and RPS23, whose expression was decreased in del(5q) when compared to non-del(5q) cells at sample level (Fig.2D), further confirming our del(5q) cell classification. Once the classification was performed, we applied a Wilcoxon signed-rank test between cells classified as del(5q) and non-del(5q), revealing in the underexpressed fraction of the genes an enrichment for the genes located on the deleted locus (Supplementary Fig.3A, B). To further validate the classification, we randomly shuffled the labels from the classified cells, and repeated the same differential expression analysis, revealing how the genes located on the deleted region started to fade away (Supplementary Fig.3C). Based on this classification, interestingly, for each individual patient, the proportion of del(5q) in the CD34+ progenitor cells was consistent with that obtained by karyotype in total bone marrow (Fig.2E).

A Heatmap of the results of CopyKat showing the copy number alteration score given to each 200kb bins in chromosome 5. In order to represent cells, a clustering has been performed within each sample (kmeans with k=80), and a posterior clustering has been applied to detect the clusters containing the cells harboring the deletion. The control sample used by the algorithm is an MDS sample with normal karyotype, while the healthy sample with normal karyotype represents an additional negative control for the analysis. B Barplot representing the percentage of cells inferred by CaSpER that harbor an amplification, a deletion or a normal number of copy number variation in each branch of chromosome 5 per patient. The control corresponds to an MDS sample with normal karyotype, which is used as a reference by the algorithm. C Venn diagram representing the number and percentage of cells classified as del(5q) by both algorithms. D Pseudobulk normalized expression of the 6 CDR-genes with higher expression in our dataset (CD74, RPS14, BTF3, COX7C, HINT1 and RPS23) separated by genotype. N=4 biologically independent samples were used. The number of del(5q) and non-del(5q) cells were used to generate the pseudobulks for each patient can be found in the Source Data. E Graph depicting the percentages of del(5q) cells inferred by karyotype, CaSpER and CopyKat for each patient. Selected cells correspond to the cells classified as del(5q) cells by both computational algorithms.

We then interrogated the distribution of del(5q) cells across the different hematopoietic progenitors. Cells with the deletion were detected in all the defined hematopoietic progenitor clusters (Fig.3A), although a high heterogeneity of distribution was observed among patients (Fig.3B, C). Despite the observed heterogeneity, a statistically significant accumulation of del(5q) cells was detected in early erythroid progenitors across all individuals (hypergeometric test, p-value<0.05). Additionally, three out of four patients exhibited statistically significant enrichment of del(5q) in granulocyte-monocyte progenitors (GMP), megakaryocyte and late erythroid progenitors (Fig.3D). Collectively, these results indicated a bias of del(5q) cells towards specific myeloid compartments, mainly towards erythroid cells, which is consistent with the association between this genetic lesion and the anemia that characterizes patients with del(5q) MDS.

A UMAP representing all the MDS samples integrated and colored by cell type. The density map represents the distribution of the cells classified as del(5q) by the two algorithms. B UMAPs with density maps representing the distribution of del(5q) cells per individual patient. C Barplots showing the number of del(5q) and non-del(5q) cells composing each cell type for each MDS patient. D Heatmap representing the enrichment of del(5q) cells (-log10(p-value)) in each cell type. Any color different from white represents a statistically significant enrichment of del(5q) cells (p-value<0.05). p-values were calculated using the one-sided hypergeometric test. The number of biologically independent replicates (cells) used for the hypergeometric test and the exact enrichment p-values can be found in the Source Data.

To delve into the transcriptional program associated with del(5q) cells in patients with MDS, we performed a pseudobulk differential expression (DE) analysis between del(5q) and non-del(5q) cells for each cell population, as traditional Wilcoxon signed-rank test -based DE analysis in single cell data has recently shown to yield high false positive rates25. Intriguingly, considering every type of hematopoietic progenitors, only seven genes were differentially expressed (downregulated) in del(5q) in comparison with non-del(5q) cells (Fig.4A). Some of the downregulated genes played a key role in MDS and other non-hematological tumors, such as PRSS21, which encodes for a tumor suppressor frequently hypermethylated in cancer26, MAP3K7CL, whose downregulation serves as a biomarker in other types of cancer27, and CCL5, whose downregulation is associated with high-risk MDS28.

A Heatmap representing the differentially expressed genes (BenjaminiHochberg-adjusted p-values<0.05 and |logFC|>2) between del(5q) and non-del(5q) cells within each hematopoietic progenitor. The heatmap was created by combining n=4 del(5q) MDS patients and generating pseudobulks per cell type. The two-sided edgeRs Likelihood Ratio Test was used to calculate p-values. The exact number of biologically independent replicates (del(5q) and non-del(5q) progenitor cells), as well as the specific p-values for each differentially expressed gene can be found in the Source Data. B Dotplot representing statistically significant biological processes and pathways (BenjaminiHochberg-adjusted p-values<0.05) for differentially expressed genes obtained in del(5q) versus Healthy and the non-del(5q) versus Healthy contrasts. The one-sided hypergeometric test was used to calculate p-values. Del(5q) and non-del(5q) cells were derived from n=4 del(5q) MDS patients, whereas healthy cells were derived from n=3 healthy donors. Biologically independent replicates (del(5q), non-del(5q) and healthy progenitor cells) used for each comparison are specified in the Source Data. CE Histograms representing the activity score in all the cells separated by conditions. Some regulons behaved similarly in the MDS samples (non-del(5q) and del(5q) cells) compared to healthy cells (C), while other regulons behaved differently in the three different conditions (D). Some inferred regulons had an activity score on the MDS samples, while lacking on the healthy samples (E).

Due to the unexpected transcriptional similarity between del(5q) and non-del(5q) cells within MDS patients, we next performed a DE analysis between del(5q) MDS cells and CD34+ cells from healthy donors. This comparison yielded 20 to 988 differentially expressed genes (FDR<0.05 and |logFC|>2), depending on the progenitor cell (Supplementary Fig.4A, Supplementary Data2). Although most of these genes were cell-type-exclusive, they were enriched in similar pathways in most of the cell types (Fig.4B, left panel). Genes overexpressed in del(5q) cells were enriched in cell cycle and mitosis-related signatures, such as DNA replication and mitotic nuclear division, and showed increased expression of DNA repair related genes, suggesting that loss of 5q confers increased proliferative potential. Additionally, del(5q) erythroid progenitors, LMPPs, GMPs, DCs, and monocyte progenitors showed significant upregulation of genes involved in the p53 signaling and, genes involved in the apoptosis pathway were significantly upregulated in LMPP, MEP and late erythroid progenitors, but not in early erythroid progenitor cells. Our results are in line with the increased levels of apoptosis described for del(5q) patients29,30,31. Downregulated genes showed enrichment in ribosomes and translation related pathways in all hematopoietic progenitors, in line with previous works that have described del(5q) MDS as a ribosomopathy2,30,32. Interestingly, besides the cytoplasmic translation, we also observed altered expression of genes associated with mitochondrial translation altered in HSCs, GMPs and granulocyte progenitors. The comparison of non-del(5q) and healthy cells resulted in 64-736 altered genes per progenitor (Supplementary Fig.4B, Supplementary Data2). Enriched processes were also homogeneous among most hematopoietic progenitors and, as expected, were similar to the ones observed in del(5q) vs healthy comparison (Fig.4B, right panel).

Despite the low number of DE genes between del(5q) and non-del(5q) cells, we were interested in understanding whether differences in GRN might be observed between these two populations. Unlike DE analysis, which is performed in a gene-by-gene manner, GRN studies use data-driven grouping of genes to enable the identification of mechanistic transcriptional differences between conditions. Thus, we applied SimiC33 to compute the regulatory activity of regulons and observed that although some regulons behaved uniformly (low regulatory dissimilarity score, in Supplementary Fig.5A black-purple color) between the three conditions (del(5q), non-del(5q) and healthy cells), a group of regulons showed differential activity (high regulatory dissimilarity score, in Supplementary Fig.5A yellow-orange color) across the conditions. Among them, three different regulon activity patterns arose. Firstly, a group of regulons that showed similar activity between non-del(5q) and del(5q) cells, and different to healthy cells, in line with DE analyses, such as the ones driven by ZNF451, YBX1 and PSPC1 (Fig.4C). Secondly, there were regulons with differential activity between the three conditions, such as those driven by JARID2, IRF1 and KAT6B, among others (Fig.4D). The three regulons showed high activity in cells from healthy age-matched controls (6184 years), whereas they presented a progressively lower activity in non-del(5q) cells, and their lowest activity in del(5q) cells. JARID2 acts as a tumor suppressor and plays a crucial role in the leukemic transformation of myeloid neoplasms34, and its deletion promotes an ineffective hematopoietic differentiation35, suggesting that the low activity of this regulon may negatively impact the hematopoietic differentiation of these patients. IRF1 is located in 5q31.1 and its deletion in one or both alleles has been observed in MDS and AML patients with chromosome 5 abnormalities36. IRF1 has been described as a master HSC regulator, and its loss impairs HSC self-renewal and increases stress-induced cell cycle activation, suggesting that its low activity in patients could confer proliferative advantage37. Decreased expression of KAT6B in aged hematopoietic stem cells has been associated with impaired myeloid differentiation38, suggesting that its almost non-existent activity in del(5q) cells may contribute to aberrant differentiation of these cells. Lastly, we detected regulons exhibiting differential activity between del(5q) and non-del(5q) cells and that showed no activity in healthy cells. In particular, regulons driven by RERE and KDM2A showed higher activity in del(5q) cells than in non-del(5q) cells (Fig.4E). RERE negatively regulates the expression of target genes, and such genes are enriched in cytoplasmic translation, ribosome biogenesis and ribonucleoprotein complex biogenesis pathways, among others (Supplementary Fig.5B). The KDM2A regulon was enriched in protein stabilization, regulation of cellular protein catabolic process and regulation of protein stability (Supplementary Fig.5B). The association of KDM2A and ribosomal genes has been already described by previous studies, postulating that KDM2A overexpression reduces the transcription of rRNA39,40.

Altogether, our results suggest a low transcriptional impact of 5q loss, with del(5q) and non-del(5q) cells presenting very similar gene expression alterations when compared to healthy controls, with such alterations being involved in processes that could contribute to abnormal hematopoietic differentiation. Nevertheless, although limited in number, genes and regulons specifically altered in del(5q) cells, such as those driven by JARID2, KAT6B, RERE or KDM2A, seem to be relevant for proliferation and myeloid differentiation, supporting the concept that cells harboring the deletion may have a more prominent role in the promotion of altered hematopoiesis.

To investigate whether the 5q deletion has a detrimental effect on cell-cell interactions between CD34+ progenitors, thus contributing to disease development, we performed a cell-to-cell communication analysis using Liana41 in both del(5q) and healthy controls datasets. We identified 4,534 interactions in healthy controls, and 314 interactions that were common to all del(5q) MDS patients, most of them overlapping with those found in healthy cells (Fig.5A). Despite this strong overlap, several differences between del(5q) MDS and healthy individuals were detected: in patients, monocyte progenitors were the most communicative cells, interacting mainly with early erythroid progenitors (Fig.5B). However, in healthy donors, HSCs, GMPs, DC, monocyte and granulocyte progenitors were the most interactive compartments, with a notable communicative pattern between granulocyte and GMP/DC progenitors (Fig.5C). Furthermore, genes involved in these differential interactions were overrepresented in different biological processes in each phenotype. For instance, interactions driven by healthy hematopoietic progenitors were enriched in negative regulation of apoptosis, HSC proliferation, leukocyte/DC differentiation, and hemopoiesis, whereas those found in MDS progenitors were enriched in negative regulation of translation, oncogenic MAPK signaling and HIF-1 signaling (Fig.5D). Focusing on interactions driven by del(5q) and non-del(5q) cells within the patients (Fig.5B), we observed very subtle differences regarding the communicational pattern and the number of interactions observed for each of the compartments, and there were no interactions specifically established between del(5q) cells, corroborating the high similarity already described between del(5q) and non-del(5q) cells. Overall, our results are consistent with the previously described lack of significant differences in gene expression between del(5q) and non-del(5q) cells, suggesting that deregulation of hematopoiesis in patients with 5q MDS affects all CD34+ cells.

A Venn diagram showing the number of unique interactions in del(5q) MDS and healthy samples. Healthy unique interactions were considered as those present in at least one of the healthy individuals, while MDS unique interactions were those that were present in all the patients. Interactions were inferred from n=4 del(5q) MDS patients and n=3 healthy donors. B Heatmap depicting the number of interactions triggered by del(5q) and non-del(5q) MDS cells, C as well as those established among healthy hematopoietic progenitors. The Source represents the cell types that express the ligand, whereas the Target represents the cells that express the receptor. D Dotplot representing statistically significant biological processes and pathways (BenjaminiHochberg-adjusted p-value<0.05) in which are enriched the encoding genes taking part in the healthy and MDS interactions. The one-sided hypergeometric test was used to calculate p-values, whose exact values can be found in the Source Data. E Chord diagram representing the unique MDS interaction AGTRAP-RACK1 among different del(5q) and non-del(5q) progenitors. F Chord diagram depicting the unique healthy interaction HMGB1-CXCR4 established by healthy hematopoietic progenitors.

To uncover specific interactions that may contribute to the disease, we next focused on those interactions that had been gained or lost in MDS versus controls. There were 17 interactions identified in patients that were totally absent in healthy individuals, suggesting that additional communications arise when developing the disease. For instance, AGTRAP expressed in monocyte and late erythroid progenitors interacted with RACK1 in HSCs, LMPPs, MEPs, pro-B and basophil progenitors (Fig.5E). AGTRAP is known to be implicated in hematopoietic cell proliferation and survival42, whereas RACK1 has been postulated as a potential therapeutic target for promoting proliferation in other myeloid neoplasms43,44. The fact that these molecules are highly expressed in MDS could potentially be contributing to the enhanced proliferation observed in MDS cells. In contrast, there were 37 interactions that appeared in the healthy donors and were absent in the patients, including the one established between HMGB1 expressed in CLPs, DC, granulocyte, basophil, megakaryocyte, early erythroid and late erythroid progenitors, and CXCR4 present in HSCs (Fig.5F). HMGB1-CXCR4 interaction is known to trigger the recruitment and activation of inflammatory cells in tissue regeneration45,46, thus its loss could have a negative impact on the bone marrow niche. In summary, these analyses may allow the identification of potential interactions implicated in the pathogenesis of the disease that could represent new therapeutic targets.

We next aimed to understand the effect of treatment with the standard-of-care, lenalidomide, on the transcriptional alterations observed in del(5q) and non-del(5q) cells. We performed scRNA-seq on CD34+ cells of two patients (Patient_5-6), which had achieved hematological response (one with partial cytogenetic response (PCR), and the other one with complete cytogenetic response (CCR), respectively) (clinical information in Supplementary Table1). Data were integrated, clustered, manually annotated, and del(5q) cells were identified as described before (Fig.6A, B). Patients showed different percentages of del(5q) cells which were consistent with karyotype results (Fig.6C): the patient with PCR showed 1939 cells with del(5q) (37.13%), whereas the patient showing CCR presented only 11 cells with del(5q) after treatment (0.15%), validating the persistence of del(5q) progenitor cells at the time of complete clinical and cytogenetic remission9. Similar to what we observed at diagnosis, the distribution of del(5q) cells was heterogeneous among patients (Fig.6A, B), and both responders exhibited a statistically significant del(5q) enrichment in GMPs and erythroid progenitors. Interestingly, patient with PCR also showed an enrichment in LMPPs, megakaryocyte, monocyte, and granulocyte progenitors (Fig.6D).

A UMAP depicting the del(5q) density across the different hematopoietic progenitors obtained in three patients after lenalidomide treatment. HSC hematopoietic stem cells, LMPP lymphoid-primed multipotent progenitors, GMP granulocyte-monocyte progenitors; granulocyte progenitors; monocyte progenitors; dendritic cell progenitors, CLP common lymphoid progenitors; B-cell progenitors; T-cell progenitors, MEP megakaryocyte-erythroid progenitors, MK_Prog megakaryocyte progenitors; early erythroid progenitors; late erythroid progenitors; basophil progenitors. B Barplots showing the number of del(5q) and non-del(5q) cells composing each cell type for each MDS patient. C Percentage of the cells identified as del(5q) by karyotype, CASPER, CopyKat, and the selection by intersecting the two algorithms. D Heatmap representing the enrichment of del(5q) cells (log10(p-value)) in each cell type. Any color different from white represents a statistically significant enrichment of del(5q) cells (p-value<0.05). P-values were calculated using the one-sided hypergeometric test. The number of biologically independent replicates (cells) used for the hypergeometric test and the exact enrichment p-values can be found in the Source Data.

We have demonstrated in the previous analyses that at diagnosis both del(5q) and non-del(5q) progenitors displayed transcriptional profiles linked to an aberrant hematopoiesis. Since both PCR and CCR patients were in hematological response, we hypothesized that the remaining CD34+ cells after lenalidomide treatment, which are mainly composed of non-del(5q) progenitors, must be able to promote improved hematopoiesis and thus restore the transcriptional profile of normal progenitor cells. To demonstrate that lenalidomide, besides the potential apoptosis of del(5q) cells, could reverse transcriptional alterations harbored by non-del(5q) cells in responder patients, we performed a DE analysis between non-del(5q) cells from the CCR and PCR and the four patients at diagnosis. This comparison revealed significant transcriptional changes after lenalidomide treatment, resulting in 6223609 genes in the different progenitor populations for the CCR (Supplementary Fig.4C, Supplementary Data2) and between 4582409 genes for the PCR (FDR<0.05 and |logFC|>2) (Supplementary Fig.4D, Supplementary Data2). Note that these transcriptional differences are significantly greater than those related to patient heterogeneity at diagnosis (see previous sections), indicating that most uncovered altered genes after treatment are probably due to treatment effect rather than patient heterogeneity. Genes altered upon treatment were enriched in ubiquitination and proteasome-mediated catabolic processes, and in phosphatidylinositol related pathways, which is in line with the mechanism of action described for lenalidomide in MDS patients47,48. Moreover, we detected an enrichment in autophagy-related processes. Overall, our results suggested an increase of the two most important protein degradation pathways in non-del(5q) cells upon lenalidomide treatment (Fig.7A, first and second panels, Supplementary Table2). Furthermore, hematopoietic progenitors exhibited an increased expression of genes involved in erythroid differentiation and erythropoietin signaling after treatment (Fig.7B), validating the enhanced erythropoiesis in response to treatment19. Our analyses also detected a positive enrichment of PD-L1 expression and PD-1 checkpoint in non-del(5q) cells of the responder patients after treatment, suggesting a potential immunosuppressive mechanism of these cells in response to lenalidomide (Fig.7A, first and second panels, Supplementary Table2).

A Dotplot representing statistically significant biological processes and pathways (BenjaminiHochberg-adjusted p-value<0.05 and |logFC|>2) for differentially expressed genes obtained in different comparisons: non-del(5q) cells of the complete responder vs at diagnosis (1st panel); non-del(5q) cells of the partial responder vs at diagnosis (2nd panel); del(5q) cells of the partial responder vs the non-responder (3rd panel); non-del(5q) cells of the complete responder vs healthy cells (4th panel); non-del(5q) cells of the partial responder vs healthy cells (5th panel). For p-value calculation, one-sided hypergeometric test was used. Specific p-values for statistically significant biological processes can be found in the Source Data. The detailed breakdown of the grouped processes shown can be found in Supplementary Table2. B Boxplot showing the normalized expression of erythroid differentiation-related genes for non-del(5q) cells in MDS at diagnosis (n=4) or after treatment with lenalidomide (partial responder, n=1; complete responder, n=1). Biologically independent replicates (cells) for hematopoietic progenitors were: Early Erythroid: n=5196 (MDS at diagnosis); n=2200 (Partial Responder); n=3496 (Complete Responder); Late Erythroid: n=7168 (MDS at diagnosis); n=1056 (Partial Responder); n=1880 (Complete Responder); MEP: n=3108 (MDS at diagnosis); n=824 (Partial Responder); n=844 (Complete Responder). Two-sided Wilcoxon signed-rank test was used to calculate p-values, that were then BenjaminiHochberg-adjusted. Box plots indicate median (middle line), 25th, 75th percentile (box) and 5th and 95th percentile (whiskers) as well as outliers (single points). Exact p-values are shown within the figure. C Graphs representing the activity scores of proliferation and differentiation-associated transcription factors in healthy cells and non-del(5q) cells of MDS patients at diagnosis and after lenalidomide treatment, as well as D in del(5q) cells of MDS patients at diagnosis, with a partial response and with no response to lenalidomide. Specific activity scores can be found in the Source Data.

GRN analyses evidenced that some of the alterations described at diagnosis were potentially reverted after treatment in non-del(5q) cells. IRF1, the master HSC regulator located in 5q31.1, which showed abnormally low activity at diagnosis in non-del(5q) cells, showed an increased activity after treatment in both patients, with the PCR not reaching the activity level seen for healthy cells, and the CCR showing an augmented activity comparable to the healthy cells (Fig.7C). KAT6B, whose lower expression has been associated with impaired myeloid differentiation, showed an augmented activity in both patients despite not reaching the activity level of healthy cells. Finally, CUX149, a TF frequently mutated in myeloid malignancies and whose knockdown leads to an MDS-like phenotype, presented similar activity in non-del(5q) cells, showing higher activity than at diagnosis (Fig.7C).

Importantly, although some transcriptional lesions were reverted upon lenalidomide treatment, non-del(5q) cells continue exhibiting altered expression of ribosome-related genes, showing a negative enrichment of processes related to ribosomes, translation, and mitochondrial translation when compared to healthy cells (Fig.7A, fourth and fifth panels, Supplementary Fig.4E, F, Supplementary Data2 and Supplementary Table2). After treatment, early and late erythroid non-del(5q) progenitors from responding patients showed no statistically significant changes in these pathways. Moreover, GRN analyses detected groups of regulons with similar activity for non-del(5q) cells at diagnosis and after treatment response, but with a different activity to the healthy cells, indicating that lenalidomide did not affect their aberrant activity. Some examples included the tumor suppressor JARID234,35, ZNF451, a TF whose high expression in leukemic cells has been associated with poor outcome50, and NCOR1, a regulator of erythroid differentiation51 (Fig.7C). Moreover, non-del(5q) cells exhibited, both at diagnosis and after treatment, abnormal high activity of two regulons that were not active in healthy cells: ADNP and SMARCE1 (Fig.7C). Globally, these results indicate that treatment with lenalidomide has the potential to revert some of the transcriptional alterations present at diagnosis in non-del(5q) cells at least in patients that responded to lenalidomide. Nevertheless, some of the transcriptional alterations present at diagnosis were not modified which could be relevant for abnormal hematopoiesis, and potentially, for the future relapse of the patients.

In line with what has been observed in non-del(5q) cells from the PCR, the remaining del(5q) cells generally exhibited significant upregulation of genes involved in ubiquitin and phosphatidylinositol signaling and, autophagy and apoptosis pathways when compared to del(5q) cells at diagnosis (Supplementary Figs.4G,6A, Supplementary Data2 and Table3), which is consistent with the mechanism of action of lenalidomide47,48. However, these cells showed reduced expression of genes implicated in ribosomal and mitochondrial translation compared to diagnosis, along with diminished expression of DNA repair associated genes. (Supplementary Fig.6A). This suggests that lenalidomidedoes not fully reverse key transcriptional alterations that may underlie the ribosomopathy characterizing the disease.

Finally, to understand the transcriptional alterations associated with a lack of hematological response after lenalidomide treatment, we performed scRNAseq on CD34+ cells of an additional patient (Patient_7), who was refractory to lenalidomide (non-responder, NR). Data were processed as described previously (clinical information in Supplementary Table1), showing 83.8% of del(5q) cells (Fig.6AC), with a statistically significant increased abundance in LMPPs, MEPs and megakaryocyte progenitors (Fig.6D). We then analyzed the transcriptional differences between the remaining del(5q) cells of the responder that presented PCR, and those of the NR patient. This analysis identified 1162244 differentially expressed genes (FDR<0.05) per progenitor (Supplementary Fig.4H, Supplementary Data2). Del(5q) cells from the patient in PCR showed statistically significant enrichment in processes and pathways related to protein ubiquitination, proteasomal protein catabolic process, phosphatidylinositol and autophagosome when compared to the NR. Moreover, these cells also exhibited an increased expression of genes involved in erythropoietin signaling when compared to the cells from the NR (Fig.7A, third panel). Interestingly, these processes are similar to the ones detected for non-del(5q) cells when comparing these cells to those at diagnosis (see previous section), and have been described as a lenalidomide response in non-del(5q) MDS patients47,48. The remaining del(5q) cells from the patient in PCR also exhibited enrichment of PD-L1 expression and PD-1 checkpoint pathway when compared to the refractory patient. These analyses suggested low transcriptional alterations promoted by lenalidomide treatment in the NR patient. Accordingly, DE analysis of del(5q) cells at diagnosis and after treatment in the NR patient yielded 20121 differentially expressed genes per hematopoietic progenitor (Supplementary Fig.4I, Supplementary Data2). These few differences resulted in subtle changes in protein ubiquitination and cell cycle-related processes after treatment (Supplementary Fig.6B, Supplementary Table4), showcasing that lenalidomide did not have a high transcriptional impact on del(5q) cells of the NR.

GRN analysis demonstrated a large number of regulons that showed changes in activity after treatment in del(5q) cells from the patient in the PCR but not in the refractory patient. For example, regulons driven by IRF1, JARID2, NCOR1, and CUX1, which showed aberrant low activity at diagnosis that was partially recovered upon treatment, presented very reduced activity in the NR patient, which was lower than that observed in the PCR, and at diagnosis (Fig.7D). Collectively, these results suggest that inNR patients, lenalidomide treatment is not able to reverse part of the transcriptional lesions carried by (5q) cells, which seems to be associated with the lack of hematological response.

Link:
Single-cell transcriptional profile of CD34+ hematopoietic progenitor cells from del(5q) myelodysplastic syndromes and ... - Nature.com