Leveraging the active conformation of LFA-1 as a potential target for hematological malignancies

Abstract

Current therapeutic strategies for non-Hodgkin’s lymphomas (NHLs) and chronic lymphocytic leukemia (CLL) are limited by systemic toxicity, acquired resistance, and tumor-mediated immune suppression. To address these challenges, we introduce a novel nano-liposomal (NP) delivery system that exploits the overexpressed, open-active conformation of lymphocyte function-associated antigen-1 (LFA-1) on malignant hematopoietic cells. This approach facilitates the selective delivery of Wiskott-Aldrich syndrome (WAS)-specific short interfering RNA (siRNA) to lymphoma and leukemia cells. Compared to conventional therapies, our LFA-1-targeted NP system provides a significant advantage by enhancing cell-specific delivery, thereby minimizing off-target effects and potentially reducing systemic toxicity. LFA-1-targeted nanoparticles facilitated the precise delivery of WAS siRNA, resulting in potent inhibition of cytoskeletal-mediated oncogenesis. In vitro assays using patient-derived NHL and CLL cells confirmed significant reductions in proliferation, migration, and invasive capacity. These findings were further validated in vivo, where the targeted delivery system effectively suppressed tumor growth in a human B-NHL xenograft model. This targeted delivery strategy offers a precision approach to gene silencing in hematologic malignancies, potentially improving therapeutic efficacy and reducing adverse effects compared to current non-selective treatments for hematological malignancies, including B-NHL and CLL.

Graphical abstract

An integrin-activated nanoparticle delivery system, functionalized with the antibody for active LFA-1 conformation, encapsulates WAS-siRNA to selectively target malignant hematopoietic cells, overcoming therapeutic resistance. This targeted intervention downregulated WASp expression, thereby inhibiting cytoskeleton-dependent proliferation, migration, and invasion of patient-derived B-NHL/CLL cells, ultimately achieving significant tumor suppression in vivo.

Introduction

Hematopoietic malignancies constitute 10% of diagnosed cancers. Non-Hodgkin’s lymphoma (NHL) and acute/chronic lymphocytic leukemia (ALL/CLL) are the most prevalent, accounting for ∼30% of all blood and childhood cancers respectively.¹ Current first-line treatments primarily involve chemotherapy, often supplemented with anti-proliferative drugs or drug combinations such as CHOP (cyclophosphamide, hydroxydaunorubicin, oncovin, prednisone), anti-CD20mAb, or other antibody (Ab)-conjugated toxins.² These regimens may also include chimeric antigen receptor (CAR) T cell therapy and targeted therapies including tyrosine kinase inhibitors. Despite these advances, many patients experience acquired resistance and relapse due to the rapid mutation of malignant cells.³

B-cell NHLs (B-NHLs) comprise a heterogeneous group of malignancies arising from mature B cells, characterized by diverse molecular subtypes, clinical presentations, and prognoses. B-NHLs represent approximately 60% of all NHL cases and include aggressive subtypes such as diffuse large B-cell lymphoma and indolent forms including marginal zone lymphoma (MZL) and follicular lymphoma.⁴ Advances in genomic and immunophenotypic characterization have revealed critical oncogenic drivers, including chromosomal translocations, epigenetic modifications, and aberrations in signaling pathways such as B cell receptor signaling, PI3K/AKT/mTOR, and the nuclear factor-κB pathway. Despite these insights, therapeutic resistance, relapse, and disease heterogeneity remain significant obstacles to achieving long-term remission. The prognosis for patients with refractory or relapsed disease remains suboptimal, highlighting the need for a novel and sophisticated approach to precisely target hematological malignancies to improve treatment outcomes.

In hematologic malignancies, integrin-mediated adhesion plays a central role in tumor progression, with lymphocyte function-associated antigen-1 (LFA-1/anti-β2/CD18) serving as a key regulator of these processes.^5,6 LFA-1, in its active conformational form, is highly expressed in hematopoietic malignancies and facilitates tumor cell motility, immune evasion, and survival by promoting adhesion to the tumor microenvironment.⁶ Given its critical role in cancer cell invasiveness,⁷ active LFA-1 represents a promising target for therapeutic intervention.

In this study, we introduce an innovative nanoparticle (NP)-based delivery system that exploits the open-active conformation of LFA-1 to selectively target malignant hematopoietic cells. As a proof of concept, we encapsulated Wiskott-Aldrich syndrome (WAS)-specific short interfering RNA (siRNA) within these NPs, leading to the selective and significant inhibition of proliferation, activation, and invasiveness in primary patient-derived B-NHL and CLL cells. We demonstrate here the high specificity of NPs targeted to active LFA-1 in downregulating WASp expression in malignant hematopoietic cells. This, in turn, downregulates the expression of the active form of LFA-1 under innately active conditions, establishing a potential bidirectional mechanistic link between WASp and active LFA-1 levels. Furthermore, systemic administration of WAS siRNA-loaded NPs in a mouse model of human lymphoma effectively attenuated tumor progression, underscoring the therapeutic potential of LFA-1-mediated targeted delivery. Collectively, these findings establish the active conformation of LFA-1 as a viable and highly specific target for NP-mediated therapeutic intervention, presenting a novel and biocompatible approach for the treatment of hematologic malignancies.

Results

Targeting active LFA-1 for selective elimination of malignant hematopoietic cells

Initially, to investigate whether elevated LFA-1 expression correlates with adverse clinical outcomes in hematological malignancies, we analyzed the survival data of 70 B-NHL patients from The Cancer Genome Atlas. The ITGAL signature, defined by the expression level of ITGAL (encoding LFA-1) and stratified by the cohort median, revealed that patients with ITGAL^high expression exhibited significantly poorer overall survival compared to the ITGAL^low group (Figure 1A, left panel; p = 0.014). To further evaluate potential synergistic effects, we examined the combined WAS-ITGAL signature, reflecting concurrent high expression of WAS (encoding WASp) and ITGAL. Notably, lymphoma patients with a WAS-ITGAL^high signature displayed markedly poorer survival (Figure 1A, right panel; p = 0.0011), representing a stronger prognostic association than ITGAL alone.

Figure 1.

Primary B-NHL cells exhibit elevated activation and preferentially uptake LFA-1-targeted nanoparticles

(A) Survival analyses of cancer datasets from the The Cancer Genome Atlas program show that gene signatures of (left) ITGAL^high and (right)ITGAL^highWAS^high correlate with lower survival of patients. p value was obtained using the log rank t test. (B–D) Patient donor-derived primary (B) B-NHL cells, (C) CLL cells, and (D) healthy donor PBMCs were analyzed by flow cytometry (FC) after staining with anti-CD69 and KIM127 (anti-activated LFA-1) antibodies. Percentages of positively stained cells are indicated relative to the unstained control for every donor. (E and F) Percentages of positively stained cells of (E) CD69 and (F) active LFA-1 are indicated along with naive inactivated healthy donor PBMCs as control (n = 3 each of patients and healthy donors). p values of one-way ANOVA with a Tukey’s post hoc test to evaluate interactions between the groups are presented. ∗p < 0.05; ∗∗p ≤ 0.01; ∗∗∗p ≤ 0.001. (G) (Left) Schematic representation of the structural composition and layering of the NPs, highlighting the surface modifications applied. (Middle and right) Dynamic light scatter analysis depicting the size distribution and Z-potential measurements for various NP formulations including unmodified NPs (NPs only), hyaluronan-coated NPs (NPs + HA), and hyaluronan and KIM127 dual-coated NPs. Data are presented as means ± SEM, demonstrating the distinct physicochemical properties imparted by surface modifications. (H–J) Confocal microscopy images illustrate the selective uptake of fluorescently labeled KIM127-coated NPs (150 μg) by different cell types, counterstained with Hoechst for nuclear visualization. (H) LFA-1-deficient K562 erythroleukemia cells serve as a negative control, confirming the lack of nanoparticle uptake. (I) PBMCs from a healthy donor show minimal to no NP uptake. (J) Primary patient-derived B-NHL cells exhibiting robust uptake of KIM127-coated NPs. (K) Representative confocal microscopy differential interference contrast and fluorescence (NPs) images of mixed patient-derived B-NHL and healthy donor PBMCs (n = 3) were acquired using a Zeiss LSM 510 META confocal microscope. Wide-field (upper) and single-cell images (bottom) represent at least 20 fields. Healthy donor PBMCs are indicated with blue arrows, and B-NHL cells are indicated with red arrows. (L) Primary patient-derived B-NHL cells were co-cultured with healthy donor PBMCs at a ratio of 1:1, followed by incubation with fluorescently labeled (rhodamine) anti-KIM127-coated NPs (150 μg). Samples were analyzed by FC. (Left) Size and granulation analysis of a mixed population pseudo-colored according to cell size gates, where blue indicates small cells (healthy donor PBMCs) and red indicates large primary patient-derived B-NHL cells. (Right) Fluorescence intensities of rhodamine reflecting the NP uptake shows that the cancer cells but not the healthy PBMCs take up the NPs (n = 3).

Given the critical role of LFA-1 expression in hematological malignancies, we next examined the basal activation profile of primary patient-derived malignant cells compared to peripheral blood mononuclear cells (PBMCs) from healthy donors. PBMCs were isolated from patients and healthy individuals, serving as biological control to establish baseline activation profile. PBMCs obtained from B-NHL (MZL phenotype) and CLL patients exhibited elevated expression levels of activation markers, including CD69 and LFA in the active “open” conformation, even in the absence of prior activation or stimulation (Figures 1B and 1C). In contrast, unstimulated PBMCs from healthy donors displayed minimal to no elevation of active LFA-1 or CD69 levels, indicating an innately naive profile compared to the constitutively active patient-derived malignant hematopoietic cells (Figures 1B–1D). Notably, leukemia and lymphoma cells innately express high levels of CD69 and activated LFA-1 compared to naive healthy donor PBMCs (Figures 1E and 1F; ∼3- to 4-fold increase in CD69 levels; ∼3.5- to 4.5-fold increase in active LFA-1 levels in B-NHL and CLL cells vs. healthy donor PBMCs; p ≤ 0.001). Given these findings, we sought to exploit the open-activated LFA-1 conformation as a candidate for selective targeting of malignant hematologic cells.

Selective uptake of anti-KIM127-coated nanoliposomes by B-NHL patients’ cells

As a next step, we developed a nanoliposome (NP)-based delivery system. NPs are pharmaceutically proven delivery vehicles that can encapsulate a therapeutic agent and display ligands that target cell-surface receptors.^8,9 The patient-derived cells express high levels of CD69 and the active conformation of LFA-1 (Figures 1E and 1F). Based on these results, an Ab that specifically recognizes the active conformation of LFA-1 (anti-KIM127)¹⁰ was designed as the coating for the NPs (Figure 1G). Dynamic light scatter (DLS) revealed a particle size of 138 ± 6.27 nm after coating with anti-KIM127 (Figure 1G). The KIM127Ab-coated NPs were labeled with rhodamine red and tested for their selective uptake. The LFA-1-deficient cell line¹¹ serving as a negative control showed no uptake of these NPs by confocal microscopy (Figure 1H). Moreover, naive healthy donor PBMCs also did not incorporate NPs (Figure 1I). As expected, anti-KIM127-coated NPs were efficiently internalized by B-NHL cells and were visible mainly in their cytoplasm (Figure 1J).¹¹

We next co-cultured B-NHL patient-derived lymphocytes with healthy donor PBMCs at a 1:1 ratio to quantify selective KIM127 NP uptake. Confocal microscopy revealed that KIM127-coated NPs were selectively internalized by primary B-NHL cells, with no uptake observed in healthy PBMCs (Figure 1K). Additionally, flow cytometry (FC) data demonstrated that 97.8% of the B-NHL cells took up NP compared to 5.24% NP uptake by healthy donor PBMCs (Figure 1L). These findings underscore the potential of KIM127-coated NPs as a therapeutic delivery platform, offering selective delivery to primary B-NHL cells and potentially other hematological malignancies.

WAS siRNA as a candidate cargo for the NP-based-delivery system

After establishing the delivery system, we optimized the cargo next. Leveraging RNAi for therapeutics provides precise targeting, versatility, and fewer side effects. We selected WAS as a candidate cargo based on our previous extensive research on WASp.¹² Thus, we obtained patient-derived B-NHL (MZL—indolent phenotype) PBMCs. As expected, WAS siRNA treatment substantially decreased WASp expression in WAS siRNA-treated B-NHL patient-derived cells in comparison to N.S siRNA treatment (Figure 2A). Furthermore, WAS siRNA treatment markedly impaired the migration of primary B-NHL cells (∼26-fold decrease in migration; 25 ± 2 μm vs. 3 ± 2 μm; Figure 2B; Videos S1 and S2; p ≤ 0.0001). This was further accompanied by a substantial reduction in activation and proliferation rates (Figure 2C; ∼2.5-fold decrease in CD69 levels; p ≤ 0.01) (Figure 2D; ∼55% ± 11% reduction in proliferation; p < 0.05). Malignant cells actively modulate their microenvironment to facilitate tissue invasiveness, a process that is cytoskeleton dependent and, as a result, enhances tumor progression. WAS siRNA-treated B-NHL patient cells displayed significantly abrogated extracellular matrix (ECM) degradation compared to non-specific (N.S) siRNA-treated cells (∼60% reduction in invasiveness: 64% ± 9% vs. 5% ± 2%; Figure 2E; p ≤ 0.0001), indicating a profound loss of invasive potential.

Figure 2.

Characterization of a candidate cargo to target hematopoietic malignant cells

(A) Primary B-NHL cells were transfected with 500 pmol WAS siRNA or N.S siRNA as control. Cells were analyzed 24 h post-treatment for WASp protein levels by FC analysis (n = 3). (B) Cell migration of primary B-NHL cells on ICAM-1/SDF-1α-coated cover glasses was tracked over a 30-min period, as described in materials and methods and shown in Videos S1 and S2. The mean cell displacement is shown (N.S siRNA: n = 214 cells; WAS siRNA: n = 148 cells) from three independent experiments. (C) Cell activation was measured via FC by staining with anti-CD69 antibody (n = 3 patient donors, a representative sample is shown). (D) Cell proliferation analysis, 24 h post-transfection; the mean proliferation rate is shown relative to the control (n = 3). Error bars represent the SE from the mean (n = 3, summary of three independent experiments). p values of two-tailed t test are presented. (E) Gelatin degradation assay. Primary B-NHL cells were treated with WAS or N.S siRNA and incubated overnight over green fluorescent gelatin coverslips. Following incubation, cells were fixed, permeabilized, and counterstained with phalloidin 594 (scale bar, 20 μm). Images were acquired with a Zeiss Confocal Laser Scanning Microscope (LSM 510 META). The images were analyzed using ImageJ software, and the percentage degradation was determined by the equation [(degraded area)/(cell area)] × 100. (N.S siRNA: N = 47 cells; WAS siRNA: N = 98 cells). Error bars represent the SE from the mean (n = 3, pooled data from three independent experiments). p values of two-tailed t test are presented. (F) (Left) Western blot (WB) analysis of WASp expression in primary CLL cells (n = 3) 24 h following transfection with 500 pmol of WAS siRNA or N.S siRNA as control. (Right) Graph presenting densitometric values of WAS vs. N.S siRNA-treated B-NHL cells (n = 3 patient donors). Error bars represent the SE from the mean (n = 3, pooled data from three independent experiments). (G) Cell migration on ICAM-1/SDF-1α-coated cover glasses was tracked over a 30-min period, as described in materials and methods and shown in Videos S3, S4, and S5. Mean cell displacement (N.S siRNA: N = 57 cells; WAS siRNA: N = 57 cells; healthy PBMCs: N = 22 cells) obtained from three different donors. Error bars represent the SE from the mean. p values of one-way ANOVA with a Tukey’s post hoc test to evaluate interactions between the groups are presented. Encapsulation of the cargo in anti-KIM127 NPs. (H) Freshly isolated healthy donor PBMCs (1 × 10⁶ cells/mL) were either stimulated with 2.5 ng/mL phorbol-12-myristate-13-acetate (PMA) and 1 μg/mL ionomycin (+) or left unstimulated (−) and incubated with WAS siRNA-loaded NPs or N.S siRNA NPs for 48 h. Cells were intracellularly stained with anti-WASp antibody, and fluorescence intensity of intracellular WASp is presented. (I and J) WB analysis of WASp expression in B-NHL (H, n = 3) and CLL (I, n = 4) cells 48 h post-transfection with N.S siRNA or WAS siRNA-loaded NPs (upper), with quantification using ImageJ software (lower). Values are shown as mean ± SEM; p value of two-tailed Student’s t test is shown. (K) Assessment of inflammatory cytokine production in healthy donor PBMCs treated with siRNA-loaded NPs or empty NPs or left untreated for 24 h. Graph showing representatives of PBMCs obtained from three healthy donors. Positive controls were treated with lipopolysaccharide (LPS) or phytohemagglutinin (PHA). Supernatants were collected and analyzed for human inflammatory cytokine levels using ELISA-based assays; values are shown as mean ± SEM. ∗p < 0.05; ∗∗p ≤ 0.01; ∗∗∗∗p ≤ 0.0001.

Video S1. Live migration analysis of WAS siRNA-treated RPMI8866 (B-NHL) cells

Playback speed is ten frames per second. One representative field out of four random fields is presented.

Video S2. Live migration analysis of N.S siRNA-treated RPMI8866 (B-NHL) cells

Playback speed is ten frames per second. One representative field out of four random fields is presented.

Notably, WAS siRNA treatment was extended to CLL patient cells to assess whether the silencing machinery could be adapted to other hematological malignancies. We observed a similar outcome with WAS siRNA-treated patient CLL cells exhibiting substantially reduced WASp expression levels (Figure 2F; ∼80% reduction; p ≤ 0.01). Additionally, the migratory activity of these cells was impaired to a greater extent than that of the naive healthy donor PBMCs (control) (46 ± 19.3 μm vs. 89 ± 33.3 μm, Videos S3, S4, and S5; Figure 2G; p ≤ 0.001). Collectively, WAS gene silencing abrogated key tumor progression hallmarks in patient samples of B-NHL and CLL, emphasizing the efficiency of WAS siRNA and its potential as a candidate cargo for the delivery system.

Video S3. Live migration analysis of WAS siRNA-treated primary CLL cells

Playback speed is ten frames per second. One representative field out of four random fields is presented.

Video S4. Live migration analysis of N.S siRNA-treated primary CLL cells

Playback speed is ten frames per second. One representative field out of four random fields is presented.

Video S5. Live migration analysis of healthy donor peripheral blood mononuclear cells (PBMC’s)

Playback speed is ten frames per sec. One representative field out of four random fields is presented.

Anti-KIM127 NPs target and selectively downregulate WASp in patient-derived cells

After successfully optimizing both the cargo and the NP delivery system, we encapsulated the siRNA cargo within the KIM127 NPs (materials and methods). To assess the specificity of activated cells, we isolated lymphocytes from healthy donors. In naive, unstimulated lymphocytes, treatment with either WAS siRNA or N.S siRNA-loaded NPs had no effect on WASp expression. However, in activated (phorbol-12-myristate-13-acetate [PMA]/ionomycin treated) lymphocytes, treatment with WAS siRNA-loaded NPs led to a reduction in WASp expression compared to N.S siRNA-loaded NPs (Figure 2H). This downregulation was specific to stimulated cells (Figure 2H).

Next, we incubated B-NHL cells with either WAS siRNA- or N.S siRNA-loaded NPs for 48 h. Treatment with WAS siRNA-loaded NPs resulted in a 68% ± 12% reduction in WASp expression levels compared to N.S siRNA-treated cells (Figure 2I; p ≤ 0.01). Similarly, in primary CLL cells, WAS siRNA-loaded NPs led to a 62% ± 9% decrease in WASp expression levels (Figure 2J; p ≤ 0.01). This reduction reflects the inherently active nature of these malignant B cells, which facilitates the efficient uptake of KIM127-coated NPs and demonstrates the targeted delivery efficiency of WAS siRNA.

Absence of aberrant cytokine secretion in PBMCs treated with WAS siRNA-loaded NPs

A major challenge in developing siRNA-based therapeutic strategies is the potential for unintended immune activation, which may trigger a pro-inflammatory response or cytokine storm due to either the NP formulation itself or its cargo. To evaluate the safety profile of KIM127-coated, siRNA-loaded NPs, we analyzed cytokine secretion in primary PBMCs from healthy donors. Specifically, we quantified interleukin (IL)-6, IL-10, and tumor necrosis factor (TNF)-α levels following NP treatment. Notably, cytokine levels in NP-treated PBMCs did not differ significantly from those in untreated controls and remained markedly lower than those observed in cells stimulated with phytohemagglutinin (PHA) or lipopolysaccharide (LPS) (Figure 2K; p = ns between untreated vs. NPs alone vs. siRNA-loaded NPs). These findings confirm that KIM127-coated NPs encapsulating WAS siRNA do not provoke a pro-inflammatory response, highlighting their potential safety for therapeutic applications.

Human B-NHL treated with WAS siRNA-loaded NPs attenuates tumor growth in vivo

Finally, we utilized an ectopic model of NOD/SCID mice bearing human Burkitt’s lymphoma cells to test the curative potential of the encapsulated NPs. The mice were subcutaneously injected with 3 × 10⁶ human B-NHL cells. Mice were intravenously administered 450 μg WAS siRNA or N.S siRNA-loaded-NPs every three days for six treatments once the tumors reached approximately 100 mm³ (Figure 3A). Tumors were harvested 48 h following the final treatment and subjected to functional assays. Mice receiving WAS siRNA-loaded NPs demonstrated a significant 4-fold attenuation in tumor volume in comparison to the N.S siRNA-loaded NPs group (273.8 ± 52 and 1,184.8 ± 259 mm³, respectively; Figures 3B and 3D; p ≤ 0.01), accompanied by a significantly reduced tumor growth rate of 11.7 ± 3 compared to 64.2 ± 15 of the N.S siRNA group (∼6-fold decrease in tumor growth rate; Figures 3C and 3D; p ≤ 0.01). Cell counting following tumor harvesting and single-cell dissociation revealed that the WAS siRNA-treated mice (1.78 ± 0.65 × 10⁷) had a significantly lower number of cells compared to N.S siRNA-treated mice (6.05 ± 1.58 × 10⁷; Figure 3E). Body weight monitoring throughout the treatment period revealed no significant differences between groups, indicating the absence of systemic toxicity (Figure 3F). Ex vivo biodistribution imaging demonstrated preferential accumulation of NPs within tumor tissues, with negligible signal detected in major organs, confirming minimal off-target deposition (Figure 3G). Consistently, gross morphology and size of major organs remained unaltered following NP administration, further supporting the systemic safety of the formulation (Figure 3G).

Figure 3.

Systemic administration of WAS siRNA-loaded NPs significantly reduces tumor progression in a human B-NHL in vivo mouse model

(A) Schematic presentation of the in vivo experimental timeline. (B) Tumor volume was assessed daily in SCID/NOD mice (n = 17 for N.S siRNA, n = 16 for WAS siRNA) treated with NPs containing either N.S siRNA or WAS siRNA (300 μg NPs, 1,000 pmol siRNA). Arrows indicate days of treatment administration. (C) The average tumor growth rate was calculated from the first treatment onward for all mice in each group. Values are shown as mean ± SEM; p value of two-tailed Student’s t test is shown. (D) Representative images of tumors at the conclusion of the experiment (tumors were taken from n = 4 mice from each group). (E) Tumor cell count post-extraction. Values are shown as mean ± SEM; the p value of the two-tailed Student’s t test is shown. (F) Body weight of mice treated with WAS siRNA-loaded NPs or N.S control siRNA NPs, administered every three days for a total of six injections. Body weights were monitored over the 18-day experimental period. No significant changes were observed between groups, indicating that the repeated NP dosing regimen was well tolerated and did not induce overt systemic toxicity. (G) Ex vivo biodistribution images of major organs (tumor, lymph nodes, heart, kidneys, lungs, spleen, and liver) collected after 18 days of treatment. Organs from mice receiving rhodamine-labeled NPs vs. control group show selective accumulation in tumor tissue, with minimal or no detectable fluorescence in normal organs, confirming targeted delivery. Moreover, there are no aberrant differences in the size or morphology of the organs visible after NP treatment vs. control group. Grayscale, rhodamine, and merged images are shown for comparison. (H) WASp expression level as detected by WB analysis. Cells obtained from tumors of three mice treated with WAS siRNA-loaded NPs or N.S siRNA-loaded NPs were lysed, and WASp expression was analyzed and normalized to GAPDH levels. Densitometric analysis of the bands was performed with ImageJ. Relative WASp expression level was compared to that in control-treated mice and is presented in the graph underneath. Values are shown as mean ± SEM; the p value of the two-tailed Student’s t test is shown. (I) WASp expression level as detected by intracellular staining with anti-WASp antibody and analyzed by FC. (J and K) Surface staining of the activation markers CD69 and active LFA-1 on tumor cells of WAS siRNA or N.S siRNA NP-treated mice, as analyzed by FC. (L) Cell growth analysis using an XTT-based proliferation assay of cells extracted from tumors of WAS siRNA or N.S siRNA NP-treated mice (data from n = 3 mice from each group were pooled and used to obtain the data). Values are shown as mean ± SEM; p value of two-tailed Student’s t test is shown. (M) Cell migration assay. Cells extracted from tumors of WAS siRNA or N.S siRNA NP-treated mice were seeded on a chambered coverslip pre-coated with ICAM-1. Cell migration was assessed using a Zeiss Observer Z1 inverted microscope as described in materials and methods. (Left) Graph summarizing the mean cell displacement (N.S siRNA: N = 344 cells; WAS siRNA: N = 233 cells, Videos S6 and S7) obtained from the tumors of three mice. Values are shown as mean ± SEM; p value of two-tailed Student’s t test is shown. (Right) Cell tracking analysis of five cells from a representative video of each group. Each line represents the pathway of a single cell. (N) Gelatin degradation assay. Cells extracted from WAS siRNA or N.S siRNA NP-treated mice were incubated overnight over green fluorescent gelatin coverslips (n = 3 mice for each treatment). Following incubation, cells were fixed, permeabilized, and counterstained with phalloidin 594. (Left) Representative images were acquired with a Zeiss Confocal Laser Scanning Microscope (LSM 510 META) (scale bar, 20 μm). (Right) Graph summarizing the percentage represented by the degraded area. The images were analyzed using ImageJ software, and the equation determined the percentage of degradation [(degraded area)/(cell area)] × 100 (N.S siRNA NPs: N = 117 cells; WAS siRNA NPs: N = 143 cells) obtained from three mice bearing B-NHL tumors. Values are shown as mean ± SEM (n = 3 mice from each group were used to obtain the tumor cells); p value of two-tailed Student’s t test is shown. n.s, not significant; ∗p < 0.05; ∗∗p ≤ 0.01; ∗∗∗p ≤ 0.001; ∗∗∗∗p ≤ 0.0001.

To assess the in vivo functional efficacy of WAS-targeted NPs, we analyzed WASp expression in B-NHL cells isolated from treated and control mice. WASp expression was reduced by 54.3% ± 12.3% in the B-NHL cells obtained from WAS siRNA-NP-treated mice compared to NS siRNA-NP-treated mice, determined both via western blot (WB) analysis and FC analysis (Figures 3H and 3I). WASp levels, in turn, correlated with the expression levels of activation molecules, CD69 and active conformation of LFA-1 (Figures 3J and 3K). These results imply that targeting WASp through active LFA-1, in turn, reduces the presence of the active LFA-1 conformation on the cancer cell surface. This could potentially limit the ability of cancer cells to adhere to each other, thereby inhibiting disease progression.

Tumor cells from WAS siRNA-NP treated mice also exhibited a reduction in proliferation of 72.2% ± 13.5% compared to the N.S siRNA-NP control group (Figure 3L; p < 0.05). Furthermore, there was a 7.5-fold reduction in migratory abilities in comparison to the N.S siRNA-loaded NP-treated cells (8.5 ± 1.6 μm vs. 65.8 ± 2.9 μm; (Videos S6 and S7); Figure 3M; p ≤ 0.0001). Finally, we detected a significant reduction in the ability of the tumor cells obtained from WAS siRNA-NP treated mice to degrade ECM and invade proximal tissues (∼40% reduction; 66.3% ± 5% in N.S siRNA-loaded NPs to 24.1% ± 3.5% in WAS siRNA-loaded NPs treated cells; Figure 3N; p ≤ 0.0001). Collectively, these results demonstrate selective in vivo targeting and modification of hematopoietic tumors of human origin using a new approach based on the recognition of an active integrin that is highly expressed in malignancies.

Video S6. Live migration analysis of WAS siRNA-treated primary patient-derived B-NHL cells

Playback speed is ten frames per second. One representative field out of four random fields is presented.

Video S7. Live migration analysis of N.S siRNA-treated primary patient-derived B-NHL cells

Playback speed is ten frames per second. One representative field out of four random fields is presented.

Discussion

This study introduces a novel NP-based therapeutic approach that selectively targets hematologic malignancies by exploiting the active conformation of LFA-1, a key adhesion molecule overexpressed in lymphoma and leukemia cells.⁷ Our findings demonstrate that functional targeting of active LFA-1 through Ab-labeled NPs enables highly specific drug delivery to hematological malignancies, including B-NHL and CLL. We demonstrate that anti-KIM127-coated NPs recognize the open-active conformation of LFA-1 on malignant primary cells, facilitating selective cargo uptake. siRNA downregulation of WASp effectively downregulated key cancer-associated pathways, leading to impaired proliferation, migration, and invasiveness both in vitro and in vivo. This was accompanied by a decrease in activation markers of active LFA-1 and CD69. Our results collectively imply that LFA-1 is critical for maintaining the malignant phenotype. Given that CD69 serves as a prognostic marker in hematological malignancies¹³ and has been linked to drug response modulation,¹⁴ our findings highlight the potential of LFA-1-targeted therapies to suppress tumor progression and enhance the efficacy of combination treatment therapies.

While advances in targeted therapies of hematological malignancies such as monoclonal antibodies, bispecific cell engagers, small molecule inhibitors, and CAR T cells have improved patient outcomes, resistance mechanisms and relapses continue to pose significant hurdles.¹⁵ Small-molecule inhibitors targeting tyrosine kinases, PI3K, and MEK/ERK pathways have shown clinical success,¹⁵ but their efficacy is often limited by the emergence of resistant phenotypes.¹² In this regard, our approach leverages active conformation of LFA-1 as a selective therapeutic entry point, addressing the need for a more precise and durable treatment strategy. Nanomedicine holds promise in addressing several limitations in treating hematological malignancies^16,17 by enhancing siRNA delivery to tumors, thereby enhancing selective uptake and therapeutic efficacy.

Previous studies have highlighted the potential therapeutic value of inhibiting LFA-1 activity and its interaction with its main ligand, ICAM.^18,19 While drugs like leukotoxin target LFA-1,^19,20 their non-specific induction of apoptosis raises concern about off-target effects. Leukotoxins exert their effects through pore formation and membrane disruption, triggering uncontrolled inflammation and potential systemic toxicity. Furthermore, our siRNA-loaded NPs exhibit a safe immune profile, as demonstrated by the absence of aberrant cytokine secretion in treated PBMCs. This targeted and controlled approach minimizes immune suppression while preserving immune function.^21,22 Moreover, stimulated PBMCs exhibit transient activation, characterized by lower and shorter-lived expression of active LFA-1 compared to hematological malignancies, which display sustained LFA-1 activation. This prolonged activation is likely driven by dysregulated integrin signaling and continuous exposure to microenvironmental factors, such as ICAM-1 and CXCL12, which promote immune evasion and support tumor survival within protective niches.⁷ This gives our system an added advantage by selectively targeting hematopoietic malignant cells.

Our in vivo data demonstrated that WAS siRNA indeed reduced active LFA-1 levels. Interestingly, analysis of hematologic cancer datasets revealed a correlation between high LFA-1 (ITGAL), WAS expression, and poor patient survival in lymphoma and leukemia cases (cBioPortal Oncoprint). This suggests a link between WASp expression and LFA-1 activity in hematopoietic malignant cells. One plausible explanation is that WASp degradation disrupts LFA-1 recycling, thereby decreasing the pool of active integrins on the cell surface. This effect may be particularly pronounced in malignant cells, which are highly dependent on LFA-1 signaling for survival and proliferation.^5,7,18 Thus, targeting integrins provides a unique therapeutic advantage as they undergo continuous internalization and recycling in leukocytes.^23,24 Efficient recycling of LFA-1 is essential for maintaining immune cell adhesion, migration, and activation. Mechanistically, talin, vinculin, and PIPKIγ coordinate signaling pathways that recruit Arp2/3 and WASp to sites of LFA-1 ligation, promoting localized actin polymerization.²⁵ This remodeling process is critical for integrin endocytosis and subsequent recycling. Loss of WASp function may impair this cytoskeletal remodeling, thereby disrupting LFA-1 trafficking and reducing its active form on the cell surface. The effect of WAS siRNA treatment thus appears to be selective for malignant cells, as these cells rely heavily on LFA-1-mediated survival signaling.^5,7 In contrast to normal immune cells, which can compensate for reduced integrin activity through alternative pathways, malignant cells may experience loss of adhesion, cytoskeletal dysfunction, and, ultimately, apoptosis upon disruption of LFA-1 signaling. This selective vulnerability makes LFA-1 targeting via WAS siRNA-loaded NPs a compelling therapeutic strategy (Figure 3), offering precise tumor cell elimination without widespread immune suppression.

Our study, utilizes targeting of two hematopoietic lineage-specific proteins as a potential therapeutic strategy for hematological malignancies, including B-NHL. Our findings establish active LFA-1 as a potential therapeutic target for precision nanomedicine-based interventions in hematologic malignancies. By leveraging LFA-1’s role in tumor cell adhesion and invasion, we developed a highly specific and adaptable treatment platform that is expected to have minimal toxicity. This RNAi-based strategy, combined with integrin-targeted NP delivery, represents a paradigm shift in hematologic cancer therapy, offering a novel approach to overcoming treatment resistance and improving long-term patient outcomes.²⁵

Materials and methods

Cell lines and primary cells

Blood samples from ALL, CLL, and NHL patients were collected at the Sheba Medical Center Hematological Institute by Prof. Avigdor Abraham (Clinical Director, Sheba Medical Center Hematological Institute) and Dr. Meirav Kedmi (Oncologist, Sheba Medical Center) in accordance with the declaration of Helsinki (0196-13-HMO) and approval by the Bar-Ilan University Ethics Committee. The cell lines and cell cultures were routinely tested for mycoplasma.

Ethical approval and patient recruitment

Peripheral blood samples were collected from patients diagnosed with B-NHL or CLL at Sheba Medical Center Hematological Institute. All the primary patient samples were obtained before the initiation of cytotoxic chemotherapy.

Blood collection and cell culture

Peripheral blood (5–10 mL) from B-NHL or CLL patients was collected into EDTA-coated vacutainers to prevent coagulation. Samples were obtained from fasting patients during their routine clinical visit. Blood samples were processed within 2 h of collection. PBMCs were resuspended in PBS or RPMI-1640 supplemented with 10% fetal bovine serum, 2 mM L-glutamine, 50 mg/mL penicillin, and 50 mg/mL streptomycin. Cell counts and viability were assessed using a hemocytometer with trypan blue exclusion. Only samples with >95% viability were used for downstream applications. All cells were grown at 37°C, under an atmosphere containing 5% CO₂.

PBMC isolation

Human PBMCs were isolated from whole blood of healthy donors or patients, by Ficoll gradient as previously described.²⁶

Western blot analysis

Western blot (WB) analysis was performed as previously described.²⁷

Gelatin degradation assay

Glass coverslips (10 mm) were placed in a 24-well plate and pre-coated with 0.01% poly-L-lysine (Sigma-Aldrich). The coverslips were then coated with 0.1% Oregon Green 488-conjugated gelatin (Invitrogen), crosslinked with 0.5% glutaraldehyde, and coated with 2 μg/mL SDF-1α (PeproTech). Lymphocytes were incubated overnight on the gelatin-coated coverslips. Following incubation, cells were fixed and counterstained with phalloidin 594. Images were acquired with the Zeiss Confocal Laser Scanning Microscope (LSM 510 META) under a 63× objective lens. The images were analyzed using ImageJ software, and the percentage of degradation was determined: %Degradation area = (Area of degraded gelatin/Area of corresponding cells) × 100.¹²

Proliferation assay

Lymphocyte proliferation was assayed with an XTT-based Cell Proliferation Kit according to the manufacturer’s instructions.

Lymphocyte migration assays

Approximately 1 × 10⁴ cells were seeded over chambered coverslips (LabTek) pre-coated with 6 μg/mL ICAM-1 and 2 μg/mL SDF-1α. Cells were incubated at 37°C with 5% CO₂, and differential interference contrast images were acquired with the Zeiss Observer Z1 inverted microscope every 5 s for 30 min under a 20× objective lens using Zen software. The random displacement of the cells was automatically tracked and analyzed using TrackMate plugin of ImageJ software. An individual trace was assigned to each of the analyzed cells. Playback speed is 10 frames/s. One representative field out of four random fields is presented.

Liposome preparation

mAb-coated NPs were prepared as previously described.²⁸ Briefly, multilamellar liposomes, composed of phosphatidylcholine (PC), dipalmitoyl phosphatidylethanolamine (DPPE), and cholesterol (Chol) at molar ratios of 6:2:1.9:0.1 PC:Chol:DPPE:DPPE-Red were prepared by a lipid-film method.²⁸ Particle diameters and zeta potential were measured using a Malvern Zetasizer Nano ZS (Malvern Instruments Ltd.) during the various stages of their preparation.

FC analysis

Extracellular staining

3 × 10⁵ lymphocytes were stained with CD69 PE-Cy5 or 1 μg/mL anti-active human LFA-1 Ab (purified from KIM127 hybridoma supernatant, ATCC Cat#: CRL-2838) for 30 min at 37°C, followed by Alexa Fluor 488-conjugated goat anti-mouse IgG1 secondary antibody staining (Jackson), 30 min on ice.

Intracellular staining

1 × 10⁶ cells/mL were either stimulated with 2.5ng/ml PMA and 1μg/ml ionomycin or left unstimulated and stained with 0.1 μg mouse anti-WASp H-250 Ab, on ice. Intracellular staining was performed as previously described.¹²

Reagents

Antibodies and their sources were obtained from the indicated suppliers. WB: mouse anti-WASp D1 (Santa Cruz), and mouse anti-GAPDH (Biodesign), HRP-conjugated goat anti-mouse (Jackson) secondary Ab. FC: PE-Cy5-conjugated mouse anti-human CD69 (BD Pharmingen), rabbit anti-WASp H-250 (Santa Cruz); the KIM127 supernatant was purified to obtain the anti-activated human LFA-1 Ab. Secondary Alexa 488-conjugated goat anti-mouse IgG1 Ab (Molecular Probes).

WAS-specific siRNA oligonucleotide was custom designed with the sequence: UGACUGAGUGGCUGAGUUA (Sigma-Aldrich).

In vitro incubation with WAS siRNA-loaded NPs

For WAS siRNA entrapment, the siRNA was mixed with full-length recombinant protamine (Sigma-Aldrich) at a 1:5, siRNA: protamine molar ratio in nuclease-free water and incubated for 20 min at room temperature to form a complex. Immediately before use, the lyophilized NPs (300 μg total lipids) were rehydrated by adding 0.2 mL nuclease-free water containing protamine-condensed siRNAs (1 nmol siRNA). Free siRNA molecules were separated by ultracentrifugation (Thermo Sorvall WX90); to determine the entrapment of WAS siRNA in the NPs, the concentration of siRNA was measured using the NanoDrop 2000C spectrophotometer (Thermo Scientific) at a wavelength of 260 nm. siRNA concentration was measured before ultracentrifugation (total siRNA) and in the supernatant of the NPs after ultracentrifugation (free siRNA). The entrapment efficiency of WAS siRNA in the NPs was calculated as follows:

$$\%siRNAentrapment=\left[\left(TotalsiRNA–FreesiRNA\right)/\left(TotalsiRNA\right)\right]\times 100\left(Entrapmentefficiencywasabove90\%inallcases\right)\text{.}$$

Survival analysis

70 B-NHL patients from The Cancer Genome Atlas were stratified based on ITGAL (LFA-1) and WAS (WASp) expression using the cohort median as the cutoff. Samples with expressions above the median were classified as ITGAL^high and those at or below as ITGAL^low. Similar was done for ITGAL-WAS^high and ITGAL-WAS^low signatures. The median threshold provided an unbiased, non-parametric division minimizing the effect of outliers.²⁹ For the statistical analysis log rank tests were applied to assess differences in overall survival.

Cytokine release assay

Freshly isolated healthy donor PBMCs (1 × 10⁶ cells/mL) were incubated for 24 h with either NP only as control or siRNA-loaded NP or were left untreated. After incubation, the supernatant from each sample was collected. Positive control samples were treated with PHA or LPS. The levels of the human cytokines, IL-6, IL-10, TNF-α, and interferon-γ were determined using Human Mini ELISA Development kits (PeproTech).

In vivo model

Detailed in Table 1.

Table 1.

In vivo model

In vivo conditions and study design
Mouse model	SCID/NOD mice (Jackson Laboratory)
Cell type and model	3 × 106 human Raji B-NHL cells, in PBS mixed with Matrigel (1:1 v/v ratio).
Number of positions injected within each mouse	1, subcutaneous
Number of mice per cage	Five mice per cage
Treatment	3 × 106 human Raji B-NHL cells, in PBS mixed with Matrigel (1:1 v/v ratio).
Treatment injection interval of NPs	Every 3 days for 6 injections, 300 μg NPs loaded with 1,000 pmol siRNA
Administration of nanoparticles	300 μg NPs every 3 days, intravenously once the tumors reach 100 mm3
Homing	SCID/NOD NSG or NRG female, 6- to 8-week-old mice were purchased from Envigo or Jackson Laboratory. All mice were housed in IVC caging and supplied with irradiated shredded corn cob bedding and irradiated mouse feed diet. The light-dark cycle was 12 h. The ambient temperature of each room was set at 20°C ± 1°C. The temperature inside the boxes generally remained between 22°C and 24°C. Humidity was set between 35% and 55%.
Ethics oversight	Housing and breeding of mice and experimental procedures were performed according to the guidelines of the Bar-Ilan University Animal Ethics Committee (#82-10-18).
Study design	Randomization was used to divide the animals for in vivo treatments.
Single-cell suspensions after harvesting	Tumors were dissociated to single-cell suspensions with 64 μg/mL collagenase type IV and DNAase. Ex vivo analyses were conducted on these single-cell suspensions.
Tumor measurements	Tumor size was measured daily using a digital caliper, and the following formula was used to calculate the tumor volume: Volume (mm3) = 0.5 × (major axis) × (minor axis)2. Tumor average growth rate was calculated according to the formula: Average growth rate = (Day of last treatment − Day of first treatment)/(Last day tumor size − Initial tumor size)
Imaging	CRI Maestro

Statistical analysis

The data and graphs were processed using Microsoft Excel (v.14.0.7) and GraphPad Prism9 and are presented as mean ± SEM. The sample size was not predetermined. Data exclusion: no data points were excluded from the final analyses. All experiments were repeated, with consistent results, and n values are included in the figure legends. Experimental groups were assigned randomly, and investigators conducting data collection, analysis, and interpretation were blinded to group allocations. The means of the two sample groups were compared by using an unpaired 2-tailed Student’s t test. The interaction value between different treatment groups was analyzed by one-way ANOVA and Tukey’s post hoc multiple comparison. A p value less than 0.05 was regarded as statistically significant. The p values along with the number of experimental repeats (n) are indicated in the figures.

Data and code availability

The datasets generated and/or analyzed and all other relevant data during the current study are available from the corresponding author upon reasonable request.

Acknowledgments

The authors thank Dr. Itay Lazar from Bar-Ilan University, and Dr. Liron Miller and Mr. Hekmatian Nassir from the blood bank and transfusion center at Sheba Medical Center, Israel, for their technical assistance. We thank Dr. Jennifer Benichou Israel Cohen for her help with the statistical analysis. The project was partially funded by the Ministry of Science and Innovation, Israel Innovation Authority.

Author contributions

Conceptualization, G.B., A.P., and M.B.-S.; methodology, G.B., A.P., B.S., M.K., A.A., and N.J.; in vivo experiments, G.B. and A.P.; blood samples, M.K. and A.A.; validation, A.P., G.B., and N.J.; formal statistical analysis, G.B., A.P., and N.J.; investigation, M.B.-S., G.B., and N.J.; data curation, A.P., G.B., B.S., and N.J.; writing – original draft preparation, A.P., G.B., and M.B.-S.; visualization and graphics, G.B. and A.P.; resources, writing – review & editing, supervision, project administration, and funding acquisition, M.B.-S.

Declaration of interests

The authors declare no potential conflicts of interest.