Anti-cd33 nanobodies, chimeric antigen receptors, and uses thereof

Abstract

The application discloses an anti-CD33 nanobody, a chimeric antigen receptor and application thereof, wherein the antibody comprises a nanobody or an antigen binding fragment thereof specifically binding to a CD33 IgV domain and a nanobody or an antigen binding fragment thereof specifically binding to a CD33 IgC2 domain. ΔE2 The antibody or the chimeric antigen receptor can not only effectively recognize and eliminate tumor cells expressing full-length CD33 and CD33 splice variants, but also reduce damage to normal hematopoietic cells, improve treatment safety, maintain good in-vivo expansion capacity and anti-tumor activity in an AML immunosuppressive microenvironment, and reduce treatment-related adverse reactions, so that safe, efficient and clinically valuable CD33-targeted immunotherapy is realized.

Description

Technical Field

[0001] This invention belongs to the field of antibody technology, specifically relating to an anti-CD33 nanobody, a chimeric antigen receptor, and their applications. Background Technology

[0002] Acute myeloid leukemia (AML) is one of the most common and worst-prognostic acute leukemias in adults, with persistently high relapse and mortality rates after standard treatment (combined chemotherapy and allogeneic hematopoietic stem cell transplantation). While chimeric antigen receptor T-cell (CAR-T) therapy has achieved breakthrough success in B-cell malignancies, it faces significant and persistent biological obstacles in the field of AML. CD33 (Siglec-3) is a 67 kDa transmembrane glycoprotein belonging to the sialic acid-binding immunoglobulin-like lectin family. It is highly expressed in approximately 85-90% of AML blast cells and leukemia stem cells (LSCs), but almost entirely absent in non-hematopoietic tissues, making it a core target for AML immunotherapy. However, translating CD33-targeted CAR-T therapy into durable and safe clinical benefits requires overcoming three closely related challenges simultaneously.

[0003] First, epitope deletion caused by natural splicing variations leads to primary drug resistance. The extracellular region of CD33 consists of a distal IgV domain (encoded by exon 2) and a proximal IgC2 domain (encoded by exon 3). Currently, the epitopes recognized by the vast majority of CD33 CAR-T cells in clinical development and marketed antibody-drug conjugates (such as gemtuzumab ozogamicin) are located in the distal IgV domain. However, in AML patients, due to mRNA alternative splicing, CD33 variants lacking exon 2 (CD33 variants) can be detected in approximately 50% of cases. ΔE2 This splicing variant retains the complete IgC2 domain, transmembrane region, and intracellular signaling motif, but completely loses the IgV domain. Therefore, CAR-T cells targeting only the IgV epitope are unable to recognize and kill CD33-expressing cells. ΔE2 The presence of leukemia cells leads to primary ineffectiveness in half of patients from the start of treatment, becoming a direct source of disease progression and relapse.

[0004] Second, severe on-target / off-tumor myelotoxicity significantly narrows the therapeutic window. CD33 is not a tumor-specific antigen; it is physiologically expressed in the normal hematopoietic system, particularly in myeloid progenitor cells, monocyte precursors, and even some hematopoietic stem / progenitor cells (HSPCs) with reconstitution potential. When highly active CD33-targeted CAR-T cells are infused, they not only eliminate leukemia cells but also indiscriminately attack normal myeloid and early hematopoietic compartments, causing irreversible severe myelosuppression. Clinically, this manifests primarily as long-term granulocytopenia (neutrophils <0.5 × 10⁻⁶). 9 The high incidence of CD33-related toxicities (CRT-C) can lead to fatal infections and is often accompanied by refractory thrombocytopenia. Multiple early clinical reports of CD33-targeted CAR-T have demonstrated this dose-limiting toxicity: high treatment-related mortality makes increasing the infusion dose in pursuit of deeper remission extremely unsafe. Furthermore, this persistent disruption of the hematopoietic microenvironment creates significant challenges for subsequent bridging radical hematopoietic stem cell transplantation.

[0005] Third, the bone marrow microenvironment of AML exerts strong immunosuppression on CAR-T cells, thereby limiting their in vivo expansion and anti-tumor persistence. Unlike B-cell hematologic malignancies, the tumor microenvironment of AML is characterized by the significant accumulation of potent T-cell inhibitory factors such as transforming growth factor-β (TGF-β), interleukin-10 (IL-10), adenosine, and prostaglandin E2. Simultaneously, negative immune cell populations, such as regulatory T cells (Tregs) and myeloid-derived suppressor cells (MDSCs), are highly infiltrated. This multi-layered immunosuppressive barrier is figuratively termed the "cytokine barrier." As a result, even with high-dose CAR-T cell infusion, the expansion in peripheral blood is often limited, and cells rapidly become functionally exhausted (upregulating inhibitory receptors such as PD-1 and TIM-3, losing cytokine secretion and killing ability), failing to establish an effective therapeutic immune response. Furthermore, increasing the infusion dose simultaneously amplifies cytokine release syndrome (CRS) and immune effector cell-associated neurotoxicity syndrome (ICANS), further reducing treatment safety.

[0006] To overcome the above obstacles, existing research has proposed a variety of design strategies, but each solution can only solve a single problem or a certain aspect of the same problem, and has not yet formed a holistic and synergistic technical solution.

[0007] 1. Single-episode CAR-T targeting the proximal IgC2 domain of the membrane To overcome CD33 ΔE2To address the resulting immune escape, some research groups have developed antibodies and CAR-T cells that recognize the proximal membrane domain of IgC2 encoded by exon 3, independent of the IgV domain. For example, single-chain antibodies (scFv) that bind to the IgC2 domain, such as antibody P67.6 or humanized h2H12, have been prepared into CARs and have been shown to kill cells expressing full-length CD33 and CD33+ cells with equal efficiency. ΔE2 The study used AML cell lines and patient-derived xenograft models. However, this approach did not solve the "on-target / off-tumor" toxicity problem because the IgC2 domain is also present on the surface of normal myeloid progenitor cells and HSPCs. Compared to the effects observed in transplanted mouse models, this type of CAR-T still caused severe pancytopenia. At the same time, its CAR backbone itself was not equipped with immunomodulatory payloads and could not actively counteract inhibitory signals such as TGF-β in the AML microenvironment, thus limiting its in vivo expansion and persistence.

[0008] 2. CD33 CAR-T cells with reduced hematopoietic toxicity through affinity optimization To explore a differentiated therapeutic window between tumor killing and hematopoietic protection, several teams have used affinity engineering to screen for mutations in CD33-targeting scFvs, deliberately reducing antigen-binding affinity from nanomolar levels (KD < 5 nM) to micromolar levels (KD ≈ 1 μM or lower). The principle is that CD33 expression density on the surface of leukemia blast cells is often higher than that of normal HSPCs; low-affinity CAR-T cells preferentially kill tumor cells with high antigen density while relatively forgiving normal myeloid precursors. Preclinical data show that this strategy does indeed alleviate the inhibition of normal hematopoietic colony formation. However, the core limitation of this approach is that existing low-affinity CAR-T conformations still generally use scFvs targeting the IgV domain, thus completely failing to cover CD33. ΔE2 This subgroup results in approximately half of the patients initially being unresponsive to treatment. Furthermore, due to significantly reduced affinity, it is difficult to reach the effective killing threshold for some leukemia cells with downregulated CD33 or those in a low antigen density state, posing a risk of secondary escape.

[0009] 3. Armored CD33 CAR-T cells co-expressing immunomodulatory factors To combat microenvironmental immunosuppression, an armored CAR-T strategy has been introduced to target CD33. A typical approach involves adding immunostimulatory elements to the same CAR vector using a polycistronic structure. For example, co-expression of the dominant-negative TGF-β receptor II (TGFβRII-dn) can block the TGF-β / Smad pathway, protecting CAR-T cells from TGF-β-mediated proliferation inhibition and Treg transformation; or co-expression of cytokines such as interleukin-15 (IL-15) and interleukin-18 (IL-18) can enhance survival and expansion through "self-feeding" signaling. These armored CAR-Ts have shown better in vivo expansion and reduced exhaustion markers in AML animal models. However, their basic recognition domains also largely target IgV epitopes, thus failing to overcome CD33. ΔE2 Related primary escape. Furthermore, armor components alone do not directly address cytotoxicity to normal HSPCs; persistent bone marrow destruction persists, and hematologic toxicity can be amplified by the enhanced persistence of CAR-T.

[0010] 4. A bridging strategy between dual-episode logic gate CAR-T cells and CD33-deficient hematopoietic stem / progenitor cells. To achieve more comprehensive AML control, research has begun to introduce combined recognition logic. One approach is to construct a dual-epitope tandem CAR, which couples an IgV-targeting scFv and a juxtamembranous C2 region scFv via a connector to form a single recognition unit, creating an "OR gate" logic—the CAR is activated by binding either epitope. This design can simultaneously recognize full-length CD33 and CD33. ΔE2 In principle, this fills the epitope coverage gap. However, the strong binding properties of "OR-gated" CARs simultaneously exacerbate the attack on normal myeloid and early hematopoiesis, resulting in more pronounced therapeutic toxicity. Another strategy is to introduce a cell-engineered salvage protocol with "on-target / off-tumor" protection: first, CD33 is knocked out of the patient's own or donor's hematopoietic stem / progenitor cells using CRISPR / Cas9, followed by reinfusion and subsequent potent CD33 CAR. T-cell therapy. This approach significantly reduces myelotoxicity because normal HSCs lose CD33 expression, thus achieving tumor exemption. However, it requires complex stem cell collection, gene editing, and transplantation processes. The long-term hematopoietic reconstitution function and safety risks (such as off-target editing and clonal expansion) of edited HSCs are not fully understood. Furthermore, this strategy typically still uses IgV-targeting single-episode CAR-T cells, which still cannot effectively eliminate CD33. ΔE2 Leukemia clones; even when the C2-CAR variant is replaced, the CAR-T still lacks microenvironment resistance equipment and faces defects such as in vivo deterioration and insufficient persistence when the armor components are not installed. Summary of the Invention

[0011] The purpose of this invention is to address the problems of existing CD33-targeted therapies, such as antigen escape, high hematopoietic toxicity, and limited efficacy in the immunosuppressive microenvironment of AML, by providing a novel immune effector cell targeting CD33 and its therapeutic strategy. This immune effector cell can effectively recognize and eliminate CD33-expressing cells and CD33-expressing cells. ΔE2 It can target spliced ​​variant tumor cells and reduce damage to normal hematopoietic cells, thus improving treatment safety. At the same time, it can maintain good in vivo expansion capacity and anti-tumor activity in the immunosuppressive microenvironment of AML and reduce treatment-related adverse reactions, thereby achieving safe, efficient and clinically valuable CD33-targeted immunotherapy.

[0012] In a first aspect, the present invention provides an anti-CD33 nanobody or an antigen-binding fragment thereof, wherein the nanobody or the antigen-binding fragment thereof is selected from at least one of the following (1)-(4): (1) The amino acid sequences of the nanobody or its antigen-binding fragment CDR1, CDR2 and CDR3 are the three CDRs of the sequence shown in SEQ ID NO:1, or have 1, 2 or 3 amino acid substitutions, deletions or additions compared with the three CDRs of the sequence shown in SEQ ID NO:1; preferably, they have 1 conserved amino acid substitution. (2) The amino acid sequences of the nanobody or its antigen-binding fragment CDR1, CDR2 and CDR3 are the three CDRs of the sequence shown in SEQ ID NO:6, or have 1, 2 or 3 amino acid substitutions, deletions or additions compared with the three CDRs of the sequence shown in SEQ ID NO:6; preferably, they have 1 conserved amino acid substitution. (3) The amino acid sequences of CDR1, CDR2, and CDR3 of the nanobody or its antigen-binding fragment are as shown in SEQ ID NO:3, SEQ ID NO:4, and SEQ ID NO:5, respectively, or have 1, 2, or 3 amino acid substitutions, deletions, or additions compared to SEQ ID NO:3, SEQ ID NO:4, and SEQ ID NO:5; preferably, they have 1 conserved amino acid substitution. (4) The amino acid sequences of the nanobody or its antigen-binding fragment CDR1, CDR2, and CDR3 are as shown in SEQ ID NO:8, SEQ ID NO:9, and SEQ ID NO:10, respectively, or have 1, 2, or 3 amino acid substitutions, deletions, or additions compared to SEQ ID NO:8, SEQ ID NO:9, and SEQ ID NO:10; preferably, they have 1 conserved amino acid substitution.

[0013] In one embodiment of the present invention, the amino acid sequence of the nanobody or its antigen-binding fragment is as shown in SEQ ID NO:1, or an amino acid sequence having at least 75%, at least 80%, at least 85%, at least 90%, at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% identity with SEQ ID NO:1; or the amino acid sequence of the nanobody or its antigen-binding fragment is as shown in SEQ ID NO:6, or an amino acid sequence having at least 75%, at least 80%, at least 85%, at least 90%, at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% identity with SEQ ID NO:6.

[0014] In one embodiment of the present invention, the nanobody or its antigen-binding fragment specifically binds to the IgV domain or IgC2 domain of CD33. In one embodiment of the present invention, the nanobody or its antigen-binding fragment shown in (1) or (3) specifically binds to the IgV domain of CD33; the nanobody or its antigen-binding fragment shown in (2) or (4) specifically binds to the IgC2 domain of CD33.

[0015] In one embodiment of the present invention, the nanobody or its antigen-binding fragment may be linked, fused or combined with one or more nanobodies, antibodies or their antigen-binding fragments with different binding specificities to form bispecific, trispecific or multispecific antibody molecules or compositions; the different binding specificities include binding specificities against different antigens, different antigenic epitopes or different functional targets.

[0016] In one embodiment of the present invention, a nanobody or its antigen-binding fragment that specifically binds to the CD33 IgV domain (preferably the nanobody or its antigen-binding fragment shown in (1) or (3)) and a nanobody or its antigen-binding fragment that specifically binds to the CD33 IgC2 domain (preferably the nanobody or its antigen-binding fragment shown in (2) or (4)) can be linked by a linker peptide or constructed by other antibody engineering methods (preferably fusion expression, chemical conjugation, etc.) to form a bispecific antibody molecule.

[0017] In a further embodiment, a nanobody or its antigen-binding fragment that specifically binds to the CD33 IgV domain (preferably the nanobody or its antigen-binding fragment shown in (1) or (3)), a nanobody or its antigen-binding fragment that specifically binds to the CD33 IgC2 domain (preferably the nanobody or its antigen-binding fragment shown in (2) or (4)), and a TNF-α binding molecule (preferably an anti-TNF-α antibody, its antigen-binding fragment, or an anti-TNF-α nanobody) can be jointly constructed to form a trispecific antibody molecule, thereby simultaneously achieving specific recognition and binding to the CD33 IgV domain, the CD33 IgC2 domain, and TNF-α.

[0018] In one embodiment of the present invention, the affinity constant (KD) of the nanobody or its antigen-binding fragment for human CD33 is less than 2 × 10⁻⁶. -9 M, preferably less than 1.75 × 10 -9 M, more preferably less than 1.5 × 10 -10 M.

[0019] In one embodiment of the present invention, the nanobody or its antigen-binding fragment can specifically recognize and bind to CD33-expressing cells.

[0020] In a second aspect, the present invention provides a chimeric antigen receptor (CAR) comprising one or more antigen-binding domains targeting CD33, wherein the antigen-binding domain comprises the aforementioned nanobody or its antigen-binding fragment.

[0021] In one embodiment of the present invention, the CAR further includes at least one of the following: (b) a hinge region, preferably the hinge region includes the hinge region of the following molecules or variants thereof: CD8α, CD28, IgG1, IgG4, CD34; more preferably the CD8α hinge region, the IgG4 hinge region or variants thereof; (c) Transmembrane region, preferably comprising transmembrane domains of the following molecules or variants thereof: CD8α, CD28, CD3ζ, ICOS, 4-1BB (CD137), OX40 (CD134), PD-1, CD4, TCRα, TCRβ, CD3ε, CD3δ, CD3γ; more preferably CD8α transmembrane region, CD28 transmembrane region or variants thereof; (d) An intracellular signal transduction domain comprising at least one co-stimulatory domain and / or one intracellular signal transduction domain, wherein the co-stimulatory domain preferably comprises a co-stimulatory signal transduction domain of the following molecules or variants thereof: 4-1BB (CD137), CD28, OX40, ICOS (CD278), CD27, GITR (CD357), CD30, HVEM, DR3, CD226 (DNAM-1), 2B4 (CD244), DAP10, DAP12; more preferably 4-1BB, CD28 or a combination thereof; wherein the intracellular signal transduction domain preferably comprises an intracellular signal transduction domain of the following molecules or variants thereof: CD3ζ, FcRγ, DAP10, DAP12, CD3ε, CD3γ, CD3δ; more preferably CD3ζ.

[0022] In one embodiment of the present invention, the CAR further comprises a signal peptide sequence, wherein the signal peptide is preferably an Igκ chain signal peptide, an immunoglobulin heavy chain signal peptide, a CD8α signal peptide, a CD4 signal peptide, a CD5 signal peptide, a CD28 signal peptide, a T cell receptor α chain (TCRα) signal peptide, a T cell receptor β chain (TCRβ) signal peptide, a GM-CSF signal peptide, or a functional variant thereof; preferably a CD8α signal peptide, an Igκ chain signal peptide, or a functional variant thereof.

[0023] In one embodiment of the present invention, the CAR further comprises a self-cleaving peptide, which is selected from T2A, P2A, E2A, F2A or functional variants thereof, preferably T2A or P2A.

[0024] In one embodiment of the present invention, the CAR further comprises a detection tag and / or an auxiliary functional element. The detection tag may be selected from truncated epidermal growth factor receptor (tEGFR), truncated CD34 (tCD34), truncated CD19 (tCD19), truncated CD20 (tCD20), truncated CD22 (tCD22), RQR8, or a combination thereof, preferably tEGFR. The auxiliary functional element may be selected from immune checkpoint or immune regulatory target binding molecules, cytokines, cytokine receptors, or functional fragments thereof. The immune checkpoint or immune regulation target binding molecules include antibodies, antibody fragments, or nanobodies targeting PD-1, PD-L1, CTLA-4, LAG-3, TIM-3, TIGIT, CD226, CD155, CD47, B7-H3, B7-H4, etc.; the cytokines or cytokine receptors include IL-2, IL-2 receptor, IL-7, IL-7 receptor, IL-15, IL-15 receptor, IL-12, IL-18, IL-21, CCL19, CCL21, or functional variants thereof.

[0025] In one embodiment of the present invention, the CAR is a single-target CAR, whose antigen-binding domain specifically binds to CD33.

[0026] In one embodiment of the present invention, the single-target chimeric antigen receptor is a single epitope CAR; preferably, it includes, in sequence, an optional signal peptide, an antigen-binding domain specifically binding to CD33, a hinge region, a transmembrane domain, a co-stimulatory signaling domain, and an intracellular signal transduction domain; preferably, the antigen-binding domain specifically binding to CD33 is any one of the nanobodies or its antigen-binding fragments mentioned in (1)-(4) above; the hinge region is a CD8α hinge region, the transmembrane region is a CD8α transmembrane domain, the co-stimulatory signaling domain is a 4-1BB co-stimulatory signaling domain, and the intracellular signal transduction domain is a CD3ζ intracellular signal transduction domain.

[0027] In one embodiment of the present invention, the single-target chimeric antigen receptor is a dual-epitope CAR, wherein the antigen-binding domain includes: (a1) First antigen-binding domain, which specifically binds to the CD33 IgV domain; (a2) The second antigen-binding domain, which specifically binds to the CD33 IgC2 domain.

[0028] In one embodiment of the present invention, the first antigen-binding domain and the second antigen-binding domain are constructed by tandem connection, fusion expression, co-expression, or other antibody engineering methods.

[0029] Preferably, the dual-epitope CAR comprises, sequentially connected, an optional signal peptide, a first antigen-binding domain, a linker peptide, a second antigen-binding domain, a hinge region, a transmembrane domain, a co-stimulatory signaling domain, and an intracellular signal transduction domain. Preferably, the first antigen-binding domain is the nanobody or its antigen-binding fragment described in (1) or (3) above; the second antigen-binding domain is the nanobody or its antigen-binding fragment described in (2) or (4) above; the hinge region is the CD8α hinge region; the transmembrane region is the CD8α transmembrane domain; the co-stimulatory signaling domain is the 4-1BB co-stimulatory signaling domain; and the intracellular signal transduction domain is the CD3ζ intracellular signal transduction domain.

[0030] In one embodiment of the present invention, the CAR is a multi-target chimeric antigen receptor, comprising the aforementioned single-target chimeric antigen receptor and a third antigen-binding domain, wherein the third antigen-binding domain specifically binds to immunosuppressive factors, immune checkpoint molecules, inflammatory factors, or tumor microenvironment regulators; preferably, the third antigen-binding domain specifically binds to TNF-α, PD-1, PD-L1, CTLA-4, LAG-3, TIM-3, TIGIT, CD47, B7-H3, B7-H4, or combinations thereof; more preferably, the third antigen-binding domain specifically binds to TNF-α.

[0031] In one embodiment of the present invention, the third antigen-binding domain is an antibody, antibody fragment, or nanobody that specifically binds to TNF-α, PD-1, PD-L1, CTLA-4, LAG-3, TIM-3, TIGIT, CD47, B7-H3, B7-H4, or combinations thereof; preferably, it is an antibody, antibody fragment, or nanobody that specifically binds to TNF-α (such as a secretory anti-TNF-α antibody, a TNF receptor-Fc fusion protein, an anti-TNF-α nanobody, or a TNF-α binding peptide, etc.).

[0032] Preferably, the multi-target chimeric antigen receptor comprises, in sequence, an optional signal peptide, a first antigen-binding domain, a linker peptide, a second antigen-binding domain, a hinge region, a transmembrane domain, a co-stimulatory signaling domain, an intracellular signal transduction domain, a self-cleaving peptide, and a third antigen-binding domain. Preferably, the first antigen-binding domain is the nanobody or its antigen-binding fragment described in (1) or (3) above; the second antigen-binding domain is the nanobody or its antigen-binding fragment described in (2) or (4) above; the third antigen-binding domain is an anti-TNF-α nanobody; the hinge region is a CD8α hinge region; the transmembrane region is a CD8α transmembrane domain; the co-stimulatory signaling domain is a 4-1BB co-stimulatory signaling domain; and the intracellular signal transduction domain is a CD3ζ intracellular signal transduction domain.

[0033] The present invention also provides a nucleic acid molecule that encodes the above-mentioned nanobody or its antigen-binding fragment, or chimeric antigen receptor.

[0034] In one embodiment of the present invention, the nucleic acid molecule further includes a promoter. The promoter is operatively linked to a nucleic acid sequence encoding the nanobody or its antigen-binding fragment or chimeric antigen receptor. The promoter is selected from the EF1α promoter, CMV promoter, EFS promoter, CAG promoter, CBh promoter, SFFV promoter, MSCV promoter, SV40 promoter, PGK promoter, MND promoter, or functional variants thereof.

[0035] In one embodiment of the present invention, the nucleic acid molecule further includes at least one Kozak sequence. The Kozak sequence is located downstream of the promoter and upstream of the nanobody or its antigen-binding fragment or chimeric antigen receptor nucleic acid sequence.

[0036] The present invention also provides an expression vector comprising the aforementioned nucleic acid molecule, wherein the expression vector is selected from plasmid vectors, lentiviral vectors, retroviral vectors, and non-viral vectors such as transposon systems (e.g., Sleeping Beauty, PiggyBac) or mRNA expression vectors.

[0037] The present invention also provides a host cell comprising the above-described expression vector.

[0038] In another aspect, the present invention provides an engineered immune cell that expresses the aforementioned chimeric antigen receptor.

[0039] In one embodiment of the present invention, the immune cells include, but are not limited to, αβ T cells, γδ T cells, NK cells, NKT cells, cytokine-induced killer cells (CIK), macrophages, and any combination thereof. CAR-T cells are preferred.

[0040] In one embodiment of the present invention, the immune cells may be autologous, allogeneic, or derived from induced pluripotent stem cell (iPSC) differentiation.

[0041] In one embodiment of the present invention, the immune cells have specific killing activity against cells expressing CD33 and can be expanded in vivo to exert an anti-tumor effect.

[0042] In one embodiment of the present invention, immune cells expressing dual-target CARs have a stronger inhibitory effect on the secretion of immunosuppressive factors compared with immune cells expressing single-target CARs.

[0043] In another aspect, the present invention provides a pharmaceutical composition comprising the nanobody, the chimeric antigen receptor, the nucleic acid molecule, the expression vector or host cell, the immune cell (preferably CAR-T cells), or a combination thereof, and a pharmaceutically acceptable carrier.

[0044] In one embodiment of the present invention, the pharmaceutical composition may further include other therapeutic agents, such as immune checkpoint inhibitors, cytokines, demethylating agents, etc., or may be administered in combination with TNF-α blockers in the treatment regimen to enhance the anti-tumor synergistic effect.

[0045] In another aspect, the present invention provides the use of the nanobody or antigen-binding fragment, the chimeric antigen receptor, the nucleic acid molecule, the expression vector or host cell, the immune cell, and the pharmaceutical composition in the preparation of a medicament for treating diseases expressing CD33.

[0046] In one embodiment of the present invention, the disease is leukemia, preferably a CD33-expressing myeloid malignant tumor such as acute myeloid leukemia (AML), myelodysplastic syndrome (MDS), chronic myelomonocytic leukemia (CMML), or juvenile myelomonocytic leukemia (JMML).

[0047] In one embodiment of the present invention, the drug can also be used to treat patients who have received myeloablative or non-myeloablative pretreatment, and to bridge hematopoietic stem cell transplantation (HSCT) after treatment to achieve long-term hematopoietic reconstitution and disease remission.

[0048] The beneficial effects of this invention are: Compared to existing CAR-T technologies that target a single epitope in the distal V domain of the CD33 membrane, this invention synergistically integrates a dual-epitope nanobody module targeting both the distal V and proximal C2 domains of the CD33 membrane with a TNF-α blocking module into the CAR design, constructing a novel immune effector cell that combines broad-spectrum recognition capability, microenvironmental adaptability, and safety advantages. This technology can simultaneously address key issues such as CD33 antigen heterogeneity, the immunosuppressive microenvironment of AML, and treatment-related toxicities, offering the following beneficial effects: (i) Achieving broad-spectrum recognition and killing of AML cells with different CD33 expression subtypes This invention screens and obtains nanobodies capable of specifically recognizing the proximal C2 domain of the CD33 membrane, and tandemly constructs a dual-epitope CAR with nanobodies targeting the V domain. This design enables CAR-T cells to simultaneously recognize tumor cells expressing full-length CD33 and CD33ΔE2 splice variants, thereby expanding the applicability of CD33-targeted therapy, improving coverage for different AML patient populations, and reducing the risk of primary drug resistance due to epitope loss.

[0049] (ii) Providing treatment strategies that connect with hematopoietic stem cell transplantation The dual epitope CAR of this invention can improve the clearance efficiency of leukemia cells, rapidly reduce tumor burden, and create favorable conditions for subsequent hematopoietic system reconstruction.

[0050] In a preferred embodiment, the present invention can be further combined with a hematopoietic stem cell transplantation (HSCT) treatment strategy to reconstruct the normal hematopoietic system after tumor removal. Simultaneously, an inducible safety switch module (e.g., the iCaspase9 system) can be configured to eliminate persistent CAR-T cells in the body after the expected therapeutic effect is achieved, thereby facilitating graft implantation and hematopoietic recovery. Through the above comprehensive treatment strategy, the clinical feasibility of CD33-targeted therapy can be improved.

[0051] (III) Enhancing the functional persistence of CAR-T cells in the AML microenvironment through TNF-α blockade This invention endows CAR-T cells with the ability to locally block TNF-α, for example by expressing or secreting TNF-α blocking nanobodies. This can effectively reduce the immunosuppressive effect mediated by TNF-α, improve the survival, expansion and sustained killing ability of CAR-T cells in the AML microenvironment, and enable them to maintain high anti-tumor activity even at lower infusion doses.

[0052] (iv) Helps reduce treatment-related inflammatory toxicity This invention enhances the anti-tumor activity of CAR-T cells while locally blocking TNF-α signaling, thereby reducing the TNF-α-mediated inflammatory cascade response, which helps to mitigate the risk of treatment-related inflammatory toxicity and improve treatment safety.

[0053] (v) Possesses strong advantages in engineering construction and industrial application. This invention uses nanobodies (VHH) as antigen recognition elements. Compared with traditional scFv, nanobodies consist of only a single structural domain and have the characteristics of small molecular weight, high stability, good solubility, and easy engineering modification.

[0054] In the construction of dual-epitope CARs, the two VHH domains can be directly tandemly linked via linker peptides, eliminating the problem of light chain and heavy chain mismatch, which is beneficial for improving the expression stability and structural uniformity of the CAR molecule. Simultaneously, this design simplifies the vector construction and cell preparation process, making it more conducive to subsequent large-scale production, quality control, and clinical translation applications.

[0055] In summary, this invention achieves broad-spectrum recognition, efficient killing, and enhanced microenvironmental adaptability of CD33-positive AML cells through a synergistic design of "dual epitope CD33 recognition + TNF-α microenvironment regulation," while also taking into account therapeutic safety and engineering feasibility, and has good prospects for clinical application. Attached Figure Description

[0056] Figure 1 The specificity and epitopes of antibody binding to CD33; Figure 2. Validate the flow cytometry results of CD33 expression in multiple cell lines; Figure 3 Flow cytometry results of antibody binding to multiple cell lines; Figure 4 Diagram of the construction of a monovalent CD33 chimeric antigen receptor; Figure 5-1 Diagram of the construction of a multivalent CD33 chimeric antigen receptor; Figure 5-2 A diagram illustrating the construction of a chimeric antigen receptor with multivalent CD33 and overexpression of TNFα antibody; Figure 6 Results of CD33 single-epitope CAR-T cell killing in vitro; Figure 7 Results of CD33 dual-epitope CAR-T cell killing in vitro; Figure 8 The in vitro cell-killing results of CD33 dual-epitope CAR-T blocking mTNF-α secretion; Figure 9 Results of in vitro detection of dual-target CAR-T cell-killing cytokines; Figure 10 Survival curves of CD33CAR-T in NCG-bearing mice, imaging luminescence intensity curves, and changes in CD3 levels in the blood; Figure 11 Survival curves of CD33CAR-T in NCG-bearing mice, imaging luminescence intensity curves, and changes in CD3 levels in the blood. Detailed Implementation

[0057] The technical solution of the present invention will be further described in detail below with reference to specific embodiments. It should be understood that the following embodiments are merely illustrative and explanatory of the present invention, and should not be construed as limiting the scope of protection of the present invention. All technologies implemented based on the above content of the present invention are covered within the scope of protection intended by the present invention.

[0058] Unless otherwise stated, the raw materials and reagents used in the following examples are commercially available products or can be prepared by known methods.

[0059] Terminology definition: CD33 (also known as Siglec-3) is a transmembrane glycoprotein belonging to the sialic acid-binding immunoglobulin-like lectin (Siglec) family. It is mainly expressed on the surface of myeloid cells and acute myeloid leukemia (AML) cells, and is an important therapeutic target for hematologic malignancies such as AML.

[0060] The IgV and IgC2 domains: The extracellular region of CD33 consists of two immunoglobulin-like domains: the distal IgV (immunoglobulin variable region-like) domain and the proximal IgC2 (immunoglobulin constant region 2-like) domain. These two domains differ in spatial location and function, and can each serve as independent antibody-binding epitope targets.

[0061] The "nanobody" or "single domain antibody (sdAb)" in this invention, also referred to as the VHH domain, are used interchangeably. They are derived from heavy-chain-only antibodies (HCAbs) found in camel-like animals (such as alpacas and llamas), and are the smallest naturally occurring antigen-binding fragments with a molecular weight of approximately 12-15 kDa. Unlike scFv, nanobodies consist of a single monomeric variable antibody domain (e.g., a heavy-chain variable region), which provides inherent structural advantages, including high solubility, thermal stability, and access to occult or hidden epitopes that are inaccessible to sterically hindered conventional antibodies or scFv. Typically, a single domain antibody or nanobody consists of four framework regions (FR1-FR4) and three complementarity-determining regions (CDR1-CDR3). In some embodiments, the single domain antibody of this invention may have a structure of FR1-CDR1-FR2-CDR2-FR3-CDR3-FR4. In some embodiments, the single-domain antibody is a camel-derived VHH, a humanized VHH, or a camel-derived VH (e.g., a camel-derived version of a human VH). The camel-derived VHH, humanized VHH, or camel-derived VH contains a domain or protein that is advantageous for use as a functional antigen-binding agent. These antibodies do not require a light chain variable region to bind to antigens with high affinity and specificity.

[0062] In this invention, "antigen-binding fragment" refers to a portion of the complete antibody structure or sequence.

[0063] The “CDR” mentioned in this invention refers to the complementarity-determining region or complementarity-determining cluster, which refers to the non-continuous antigen-binding sites found in the variable region.

[0064] The residues in each of these CDRs are described in the table below, depending on the different CDR determination schemes.

[0065]

[0066] When referring to antibodies defined by a specific CDR sequence as defined in this invention, the scope of said antibody also includes antibodies whose variable region sequence contains the specific CDR sequence, but whose claimed CDR boundaries differ from those defined in this invention due to the application of different schemes (e.g., different assignment system rules or combinations).

[0067] The boundaries of the CDR of the antibody of the present invention can be determined artificially according to any method or combination thereof in the art. Unless otherwise stated, in this invention, the term "CDR" or "CDR sequence" covers the CDR sequence determined in any of the foregoing methods.

[0068] In this invention, "homology" refers to a high proportion of amino acid or nucleotide matching shown when the target amino acid or nucleotide sequence is compared with a reference sequence. Homology in this invention can be determined using standard software such as BLAST or FASTA.

[0069] In this invention, "75% or more of the sameness" means that the homology is 75% or more, preferably 80% or more, more preferably 85% or more, more preferably 88% or more, further preferably 90% or more, further more preferably 93% or more, particularly preferably 95% or more, particularly more preferably 98% or more, and most preferably 100% homology.

[0070] In this invention, "specific binding" means that the antigen recognition region of this invention does not cross-react with, or substantially does not cross-react with, any polypeptide other than the target antigen. The degree of specificity can be determined using immunological techniques, including but not limited to Western blotting, immunoaffinity chromatography, and flow cytometry.

[0071] Bispecific / trispecific / multispecific antibodies are engineered antibody molecules that can simultaneously recognize and specifically bind to two, three, or more different antigens or epitopes. These antibodies can be constructed through linker peptide linkage, fusion protein expression, or other antibody engineering methods, and offer the advantage of multi-target synergistic action.

[0072] Linker peptides are short peptide sequences used to connect two or more protein functional domains, such as (G4S). n Flexible linker peptides. The length and composition of linker peptides affect the spatial conformation and relative independence of the functional domains of the fusion protein.

[0073] The "chimeric antigen receptor" (CAR) in this invention is an artificially modified receptor capable of anchoring specific molecules (such as antibodies) that recognize tumor cell surface antigens onto immune cells (such as T cells), enabling the immune cells to recognize tumor antigens or viral antigens and kill tumor cells or virus-infected cells. A CAR typically comprises, in sequence, an optional signal peptide, a polypeptide (such as a single-chain antibody) that binds to tumor cell membrane antigens, a hinge region, a transmembrane region, and an immune-active cell activation signal transduction region, wherein the immune-active cell activation signal transduction region further includes a co-stimulatory domain and a signal transduction domain.

[0074] The term "functional variant" refers to a chimeric antigen receptor, polypeptide, or protein that has significant sequence identity or similarity to its parental chimeric antigen receptor, and retains the biological activity of its parental chimeric antigen receptor. Functional variants include those that retain the ability to recognize target cells to a degree of similarity, equalness, or greater than that of the parental chimeric antigen receptor. Functional variants may have at least about 30%, 50%, 75%, 80%, 90%, 98%, or higher amino acid sequence identity relative to the parental chimeric antigen receptor.

[0075] The chimeric antigen receptors (including functional portions and functional variants) described in this invention may be glycosylated, amidated, carboxylated, phosphorylated, esterified, N-acylated, cyclized (by, for example, a disulfide bridge), or converted into acid addition salts and / or optionally dimerized or polymerized, or coupled.

[0076] "Conservative amino acid substitution" refers to replacing one amino acid with another amino acid that has similar physicochemical properties (such as charge, polarity, hydrophobicity, and size). Such substitutions generally do not affect the spatial conformation and biological function of the protein. For example, replacing leucine (Leu) with isoleucine (Ile), or replacing aspartic acid (Asp) with glutamic acid (Glu).

[0077] CAR-T cells: Chimeric antigen receptor T cells are immune cells that have been genetically engineered to express chimeric antigen receptors (CARs) on their surface. CARs typically consist of an extracellular antigen recognition domain (such as nanobodies or scFvs), a transmembrane domain, and an intracellular signal transduction domain, enabling T cells to specifically recognize and kill target cells in a non-MHC-restricted manner.

[0078] Example 1: Screening of CD33 nanobodies and determination of antibody specificity 1. Animal immunization and immune response assays 1.1 Alpaca Immunity (1) Human Siglec-3 / CD33 Protein (Kaika Biotechnology, catalog number: SIG-HM203) and CD33 Domin C2 synthetic peptide (Huakang Biotechnology) were used as immunogens to immunize alpacas. Two alpacas were immunized in total, and a total of 8 immunizations were performed.

[0079] (2) For the first immunization, 200 μg of Human Siglec-3 / CD33 Protein (dissolved in 1 mL of 1×PBS) was emulsified with 1 mL of Freund's complete adjuvant (Sigma, catalog number: F5581-10 mL) and injected subcutaneously at four points, 500 μL at each point. For the second to eighth immunizations, Freund's incomplete adjuvant (Sigma, catalog number: F5506-10 mL) was used.

[0080] (3) Second to fifth immunizations were administered at intervals of 20, 34, 48, and 62 days after the first immunization, with each booster immunization using 100 μg of Human Siglec-3 / CD33 Protein. Sixth to eighth immunizations were administered at intervals of 76, 90, and 104 days after the first immunization, with each booster immunization using 100 μg of CD33 Domin C2 synthetic peptide.

[0081] 1.2 Immune Response Assay Peripheral blood samples of 10 mL were collected 7 days after the 3rd to 7th immunizations, and 50 mL were collected 7 days after the 8th immunization. Serum samples were separated, and a portion was used for immunization titer testing.

[0082] Assess the immune status of alpacas by ELISA: (1) Dilute non-biotinylated human Siglec-3 / CD33 protein (CD3-HM433) to 1 μg / mL with coating buffer (Solepro, catalog number: C1050-500ml), and coat each well with 100 μL of the microplate (Corning, catalog number: 42592) at 4°C overnight. (2) Wash the microplate three times with 300 μL of washing solution [1×PBST (0.05% Tween 20)] per well, add 200 μL of blocking buffer [1×PBS (3% BSA)] to each well, and incubate at 37°C for 2 hours. (3) Dilute alpaca serum with diluent [1×PBST(1% BSA)] at 1 / 1000, 1 / 2000... 12 gradients. After blocking, wash the ELISA plate 3 times with detergent. Add 100 μL of diluted serum to each well and incubate at room temperature for 1 hour. (4) Dilute the antibody Gaot pAb to Llama IgG H&L (HRP) (Abcam, catalog number: ab112786) and Mouse Anti-Llama IgG2 / IgG3-HRP (SouthernBiotech, catalog number: 5880-05) at 1:2W and 1:2K respectively. Wash the ELISA plate 3 times and add 100 μL of antibody to each well. Incubate at room temperature for 30 minutes. (5) Wash the ELISA plate 3 times and add 100 μL of TMB chromogenic solution (abcam, catalog number: ab171524) to each well. Incubate at room temperature for 5 minutes. (6) Add 50 μL of stop solution (Solepro, catalog number: C1058) to each well to terminate the reaction, and measure the absorbance at 450 nm and 630 nm.

[0083] 1.3 Experimental Results After the eighth immunization, the immune titers of both alpacas remained at a high level, meeting the requirements for phage library construction.

[0084] 2. Construction of Nanobody Libraries 2.1 RNA extraction (1) Peripheral blood mononuclear cells were obtained by separating 100 mL of peripheral blood collected after the last immunization using density gradient centrifugation with human mononuclear cell separation solution (Dongfang Huahui, catalog number: 25910).

[0085] (2) Total RNA was extracted using an RNA extraction kit (Adley, catalog number: RN28). The RNA extraction method was performed according to the kit instructions.

[0086] (3) The total RNA quality was assessed by agarose gel electrophoresis and quantified by measuring the absorbance at 260 nm / 280 nm.

[0087] 2.2 Antibody Fragment Amplification (1) Using a reverse transcription kit (TAKARA, catalog number: 6210A), total RNA was reverse transcribed into cDNA using Oligo dT and Random 6 mers double primers; (2) The VHH fragment was obtained by one-step PCR and nested PCR using Prime STAR high-fidelity DNA polymerase (TAKARA, catalog number: R045).

[0088] 2.3 Construction of Escherichia coli library (1) The phage vector pComb3XSS and the purified product obtained by PCR amplification were digested with SfiI (NEB, catalog number: R0123S). The digested products were subjected to 1% and 1.5% agarose gel electrophoresis, respectively. The gel was recovered using a DNA recovery kit and ligated with T4 DNA ligase (NEB, catalog number: M0202S) at a molar ratio of 1:3. The mixture was incubated overnight at 16°C.

[0089] (2) The ligation products were purified using a DNA recovery kit.

[0090] (3) The purified ligation product was electroporated into TG1 competent cells (lucigen, catalog number: 60502-02). The specific operation is as follows: ① Take the lucigen competent cells, thaw them on ice, and add 500 ng of ligation product to each competent cell. ② Add the mixture of competent cells and ligation product to a 1 mm electrode cup and electroporate. Immediately after electroporation, add 440 μL of SOC medium preheated at 37℃. ③ Transfer the transformation product in the electroporation cup to a 50 mL centrifuge tube and activate it at 37℃ and 250 rpm for 1 hour on a constant temperature shaker. ④ Take 100 μL of the activated bacterial solution and perform serial dilutions of 1000 times, 10000 times, and 100000 times. Take 500 μL of each dilution and spread it on a plate. Spread the remaining stock solution on one plate per 1 mL and incubate overnight at 37℃. ⑤ On the second day, the bacterial cells from each plate of the original culture medium were scraped off using a disposable applicator using 1.5 mL of 2×YT medium. The resulting bacterial suspension was the constructed CD33 single-domain heavy chain antibody phage display library. By counting the bacterial cells on the serially diluted plates, the library sizes for the two alpacas were found to be 4.0 × 10⁻⁶. 8 and 1.44×10 8 .

[0091] 2.4 Document Quality Inspection To detect the insertion efficiency, packaging efficiency, and sequence diversity of the E. coli library, 30 single clones were randomly selected from each of the two alpacas for colony PCR, phage ELISA, and Sanger sequencing. The results showed that the fragment insertion rate of both alpacas was 100%, the packaging efficiency was 96.7%, and the sequence diversity was 100%.

[0092] 3. Enrichment and Screening of Nanobodies 3.1 Amplification of phage libraries (1) Take 50 μL of E. coli library and inoculate it into 100 mL of 2×YT / Amp / 1%Glu medium, and culture with shaking until the logarithmic phase. (2) Add helper phage M13KO7 at a ratio of 20 times the number of bacteria, mix gently, and incubate at 37℃ for 30 min, followed by shaking for 30 min. (3) Centrifuge the culture medium and discard the supernatant. Resuspend the precipitate in 100 mL of 2×YT / Amp / Kana medium and incubate overnight at 30℃ to amplify the phage displaying nanobodies. (4) Centrifuge the culture medium and take the supernatant. Add 1 / 5 volume of pre-cooled 20% PEG8000 / 2.5M NaCl solution to precipitate the phage. (5) Centrifuge and discard the supernatant. Resuspend the phage precipitate in 1 mL of pre-cooled 1×PBS. Centrifuge again, discard the impurity precipitate, and transfer the phage supernatant to a clean centrifuge tube. (6) Take 10 μL of phage, dilute it 10 times, measure the A260 value, and calculate the titer. The calculation formula is: dilution factor × A260 reading × 22.14 × 10 10pfu / mL.

[0093] 3.2 Phage enrichment screening Both alpacas underwent two rounds of solid-phase phage library screening. The first round used Non-biotinylated Human Siglec-3 / CD33 Protein (final concentration 5 μg / mL) for screening. The specific steps were as follows: (1) Dilute the Non-biotinylated Human Siglec-3 / CD33 Protein to 5 μg / mL with coating buffer, and coat each well with 100 μL of the microplate. Incubate overnight at 4°C. (2) Wash the microplate three times, add 200 μL of blocking buffer to each well, and incubate at 37°C for 2 hours. (3) Block with blocking buffer [1×PBS (5% skim milk powder)] 1×10 12 After blocking the PFU phage, wash the ELISA plate three times with detergent, add 100 μL of blocked phage to each well, and incubate at 37°C for 1.5 h. (4) Wash the ELISA plate six times, add 100 μL of 0.1 M Glycine (pH=2.2) to each well, and incubate with shaking for 10 min. (5) Transfer the elution product to a clean 1.5 mL centrifuge tube, add an appropriate amount of 1 M Tris·HCl (pH=8) for neutralization. (6) Add the collected phage neutralization solution to TG1 in the logarithmic growth phase, and incubate at 37°C for 30 min. (7) Dilute the phage-infected TG1 in a 10-fold gradient and plate it. Plate all the remaining bacteria as stock solution. Calculate the titer of the phage nanobody library in one round of panning based on the number of plaques on the second day.

[0094] In the second round, solid-phase panning was performed using non-biotinylated Human Siglec-3 / CD33 Protein (final concentration 2.5 μg / mL), CD33 Domin V peptide (final concentration 3 μg / mL), and CD33 Domin C2 peptide (final concentration 3 μg / mL). The resulting panned library was obtained by scraping the original culture medium from the first round of panning onto 2YT / Amp medium. This library was then infected with helper phages for the second round of panning. During the second round of panning, the phages were washed 12 times after incubation. Through multiple rounds of panning, positive clones were continuously enriched, thus achieving the goal of screening CD33-specific antibodies from the antibody library using phage display technology.

[0095] 3.3 Positive rate detection of the selected document library The positive rate of the panned library was detected by phage ELISA, similar to the method in 1.2. The primary antibody was monoclonal phage supernatant diluted 1:1 with 1×PBST (containing 5% skim milk powder). The preparation method was as follows: (1) Pick 48 single colonies from different panning rounds and culture them in deep well plates until the logarithmic growth phase. (2) Add helper phage M13KO7 at a ratio of 20 times the number of bacteria, mix gently, and incubate at 37°C for 30 min, then shake for 30 min. Add Kana antibiotic to each well and incubate at 30°C overnight. (3) Centrifuge the culture medium and collect the supernatant to obtain the monoclonal phage displaying nanobody. The secondary antibody was Anti-M13 Bacteriophage,AlpHcAbs@Rabbit antibody (HRP) (Critical Point Bio, catalog number: 052-201-005). A sample well is considered a positive clone when the OD450-OD630 value is greater than 2.5.

[0096] 3.4 Experimental Results ELISA testing showed that both alpacas had a high positive rate, making them suitable for high-throughput screening.

[0097] 4. Screening and identification of positive monoclonal antibodies 4.1 Phage ELISA 4.1.1 Experimental Methods (1) Following the experimental method in 3.3, single clones were selected from the libraries of two alpacas obtained after two rounds of screening and tested by ELISA. (2) According to the judgment principle, positive clones were subjected to Sanger sequencing. (3) The single clones that underwent Sanger sequencing were sequenced using the software SnapGene, and sequence-specific single clones were selected.

[0098] 4.1.2 Experimental Results The positive clones screened in this embodiment are sequence-specific nanobodies numbered VHH01 and VHH19, and their corresponding variable region amino acid sequences and CDR1~CDR3 sequences (IMGT scheme) are shown in the table below: the variable region nucleotide sequence of VHH01 is shown in SEQ ID NO: 2, and the variable region nucleotide sequence of VHH19 is shown in SEQ ID NO: 7.

[0099]

[0100] Antibodies targeting the CD33 antigen were enriched from a phage antibody library through affinity panning, and clones binding to the CD33 antigen were obtained through preliminary screening. However, the binding ability and specificity of the antibody molecules expressed in the prokaryotic system after conversion to IgG antibody molecules expressed in the eukaryotic system needed further confirmation. Therefore, IgG expression plasmids for these clones were prepared and expressed by transient transfection of 293 cells, and Protein A was purified into antibodies. Then, the binding specificity of the monoclonal antibodies was further determined by ELISA and flow cytometry.

[0101] 4.2 At the protein level, ELISA was used to analyze the specificity and epitopes of antibody binding to CD33. 4.2.1 Experimental Methods The binding specificity and epitopes of the two clones to the CD33 antigen protein were analyzed using ELISA, and any non-specific binding with other unrelated CD33 antigens was also analyzed. The method was similar to 1.2, except that the non-biotinylated Human Siglec-3 / CD33 Protein, Human CD5 Protein (Abcam, catalog number: CD5-HM101), Human CD27 Ligand / CD70 Trimer Protein (Abcam, catalog number: CDL-HM427), and CD33 Domin C2-His protein were diluted to 1 μg / mL with coating buffer, and 100 μL was coated per well onto the ELISA plate. The two antibody proteins and the GO positive control were used as primary antibodies, diluted to 50 nmol. The secondary antibody was GOat Anti-Human IgGFc (HRP) preadsorbed (Abcam, catalog number: ab98624).

[0102] 4.2.2 Experimental Results like Figure 1The clone GO / Anti-Human IgG Fc HRP Ab shown is a CD33 positive control antibody (gemtuzumab, GO), and Anti-Human IgG Fc HRP Ab is a negative control with only the second antibody added. Both antibodies (VHH01-hIgG1Fc and VHH19-hIgG1Fc) can bind to non-biotinylated Human Siglec-3 / CD33 Protein protein, but do not bind to two unrelated antigens (Human CD5 Protein and Human CD27 Ligand / CD70 Trimer Protein), indicating that both antibodies are clones that specifically bind to CD33. VHH01-hIgG1Fc does not bind to CD33 Domin C2-His protein, indicating that the epitope of this antibody protein is located at Domin V, while VHH19-hIgG1Fc binds to CD33 Domin C2-His protein, indicating that the epitope of this antibody is located at Domin C2.

[0103] 4.3 At the protein level, FACS analysis was used to obtain the binding specificity of the antibody to multiple cell lines. 4.3.1 Experimental Methods The binding specificity of the two clones to the overexpressing cell line was analyzed using FACS. The specific operation was as follows: (1) Take 1×10 5 (1) Place 293T cell lines, Jurkat cell lines, THP-1 cell lines, U937 cell lines, and 293T cell lines overexpressing CD33 into 1.5mL EP tubes, centrifuge at 400g for 5min, and discard the supernatant. (2) Use two antibody proteins and a GO positive control as primary antibodies, dilute to 200nm, add 100μL of the corresponding antibody diluent, resuspend the cells, and incubate on ice for 30min. (3) Centrifuge at 400g for 5min, discard the supernatant, add 1mL of 1×PBS to each tube, and resuspend the cells using a vortex mixer. (4) Centrifuge at 400g for 5min, discard the supernatant, and use diluent to prepare flow cytometry antibody APC anti-human IgG Fc Recombinant Antibody (Biolegend, catalog number: 366905) at a ratio of 1:200, add 100μL of flow cytometry antibody to each tube, resuspend the cells, and incubate on ice in the dark for 30min. (5) Centrifuge at 400g for 5 min, discard the supernatant, add 1 mL of 1×PBS to each well to resuspend the cells, repeat the operation once, centrifuge again, and add 100 μL of 1×PBS to each tube to resuspend the cells. (6) Use the APC pathway for detection.

[0104] 4.3.2 Experimental Results Flow cytometry results of cell binding were shown Figure 2-3 middle, Figure 2 To detect the results of antibody staining in this experiment, the staining results of the various cell lines used were negative with 293T and Jurkat, and positive with THP-1, U937, and 293T-CD33 cell lines. Figure 3 The following are flow cytometry results of the binding of the two cloned proteins screened in this invention to various cell lines: APC-anti-human IgG Ab is a negative control with only the second antibody added, cell only is a negative control without antibody added, and GO / APC anti-human IgG Ab is a CD33 positive control antibody (gemtuzumab ozogamicin, GO). The THP-1 and U937 cell lines naturally express the CD33 antigen, and both VHH01 and VHH19 clones can bind to it, although the binding of VHH19 is weaker than that of VHH01. The 293T-CD33 cell line overexpresses the CD33 antigen, and both VHH01 and VHH19 clones can bind to it, although the binding of VHH19 is weaker than that of VHH01. The flow cytometry results for the three CD33 antigen-negative cells, 293T and Jurkat, were all negative, indicating that the two selected cloned antibodies can specifically recognize and bind to CD33-expressing cells.

[0105] 4.4 Affinity Measurement The affinity of each clone for CD33 was determined by SPR using a BIAcore T200. Briefly, VHH-hIgG1Fc protein was covalently coupled to a ProteinA chip. Non-biotinylated Human Siglec-3 / CD33 Protein was serially diluted in running buffer (10 mM HEPES, 150 mM NaCl, 3 mM EDTA, 0.05% surfactant P20, pH 7.4) and injected at a flow rate of 30 μL / min, followed by dissociation. The binding affinity of different clones is shown in Table 1.

[0106] Table 1. Binding affinity of VHH-hIgG1Fc to human CD33-His protein

[0107] Example 2: Preparation of a monovalent CD33 chimeric antigen receptor A nucleic acid sequence encoding a backbone polypeptide comprising, from N-terminus to C-terminus, a CD8α hinge domain, a CD8 transmembrane domain (TM), a CD137 (4-1BB) cytoplasmic domain and a CD3ζ cytoplasmic domain, a 2A cleavage peptide, and EGFRt was chemically synthesized and cloned downstream of a pre-modified lentiviral vector and operatively ligated to a constitutive hEF1α promoter. The resulting CAR backbone vector was named “PCDH-EF1α-88137Z”. The multiple cloning site (MCS) in the vector allows a nucleic acid sequence containing a Kozak sequence (operatively ligated to a nucleic acid sequence encoding a CD8α signal peptide fused to the N-terminus of a VHH fragment) to be inserted upstream of the PCDH-EF1α-88137Z vector and operatively ligated to the CAR backbone sequence.

[0108] To construct a single-specific CAR with a single VHH domain using the PCDH-EF1α-88137Z backbone, a nucleic acid sequence encoding the VHH domain was operatively linked to the 3' end of a nucleic acid sequence encoding the CD8α signal peptide. The fusion nucleic acid sequence was chemically synthesized and cloned into the PCDH-EF1α-88137Z backbone using EcoRI via molecular cloning techniques known in the art. Figure 4 The vector structures constructed to express exemplary single-specific, monovalent anti-CD33 CARs are listed. The amino acid sequence of the VHH01 CAR is shown in SEQ ID NO: 11, and the nucleotide sequence of the VHH01 CAR is shown in SEQ ID NO: 12. The amino acid sequence of the VHH19 CAR is shown in SEQ ID NO: 13; the nucleotide sequence of the VHH19 CAR is shown in SEQ ID NO: 14.

[0109] Example 3: Preparation of a multivalent CD33 chimeric antigen receptor A nucleic acid sequence encoding a backbone polypeptide comprising, from N-terminus to C-terminus, a CD8α hinge domain, a CD8 transmembrane domain (TM), a CD137 (4-1BB) cytoplasmic domain and a CD3ζ cytoplasmic domain, a 2A cleavage peptide, and EGFRt was chemically synthesized and cloned downstream of a pre-modified lentiviral vector and operatively ligated to a constitutive hEF1α promoter. The resulting CAR backbone vector was named “PCDH-EF1α-88137Z”. The multiple cloning site (MCS) in the vector allows a nucleic acid sequence containing a Kozak sequence (operatively ligated to a nucleic acid sequence encoding a CD8α signal peptide fused to the N-terminus of a VHH fragment) to be inserted upstream of the PCDH-EF1α-88137Z vector and operatively ligated to the CAR backbone sequence.

[0110] To construct a single-specific CAR with two VHH domains using the PCDH-EF1α-88137Z backbone, a nucleic acid sequence encoding the two VHH domains was operatively ligated to the 3' end of a nucleic acid sequence encoding the CD8α signal peptide. The fusion nucleic acid sequence was chemically synthesized and cloned into the PCDH-EF1α-88137Z backbone using EcoRI, a molecular cloning technique known in the art. The resulting CAR backbone was named “PCDH-EF1α-VHH01-VHH19-88137Z”. To construct a single-specific CAR for a GO positive control using the PCDH-EF1α-88137Z backbone, a nucleic acid sequence encoding the scFv domain was operatively ligated to the 3' end of a nucleic acid sequence encoding the CD8α signal peptide. The fusion nucleic acid sequence was chemically synthesized and cloned into the PCDH-EF1α-88137Z backbone using EcoRI, a molecular cloning technique known in the art. Figure 5-1 The structures constructed to express exemplary single-specific, multivalent anti-CD33 CARs and GO positive control CAR structures are listed. Specifically, the amino acid sequence of the VHH01-VHH19 CAR is shown in SEQ ID NO: 15; the nucleotide sequence of the VHH01-VHH19 CAR is shown in SEQ ID NO: 16; the amino acid sequence of the VHH19-VHH01 CAR is shown in SEQ ID NO: 17; and the nucleotide sequence of the VHH19-VHH01 CAR is shown in SEQ ID NO: 18.

[0111] Example 4: Preparation of a chimeric antigen receptor with multivalent CD33 and overexpression of TNFα antibody A nucleic acid sequence encoding a backbone polypeptide comprising, from N-terminus to C-terminus, a CD8α hinge domain, a CD8 transmembrane domain (TM), a CD137 (4-1BB) cytoplasmic domain and a CD3ζ cytoplasmic domain, a 2A cleavage peptide, and EGFRt was chemically synthesized and cloned downstream of a pre-modified lentiviral vector and operatively ligated to a constitutive hEF1α promoter. The resulting CAR backbone vector was named “PCDH-EF1α-88137Z”. The multiple cloning site (MCS) in the vector allows a nucleic acid sequence containing a Kozak sequence (operatively ligated to a nucleic acid sequence encoding a CD8α signal peptide fused to the N-terminus of a VHH fragment) to be inserted upstream of the PCDH-EF1α-88137Z vector and operatively ligated to the CAR backbone sequence.

[0112] To construct a membrane-expressing TNFα antibody (adalimumab) CAR using the PCDH-EF1α-VHH01-VHH19-88137Z backbone, a nucleic acid sequence encoding adalimumab, a CD8α hinge domain, and a CD8 transmembrane domain (TM) was operatively ligated to the 3' end of a 2A cleavage peptide nucleic acid sequence. The fusion nucleic acid sequence was chemically synthesized and cloned into the PCDH-EF1α-VHH01-VHH19-88137Z backbone using SalI via molecular cloning techniques known in the art. Figure 5-2 The structures constructed to express an exemplary single-specific, multivalent anti-CD33 CAR and overexpressing an antibody against TNFα are listed. The amino acid sequence of the VHH01-VHH19-TNFαblock CAR is shown in SEQ ID NO: 19; the nucleotide sequence of the VHH01-VHH19-TNFαblock CAR is shown in SEQ ID NO: 20.

[0113] Example 5: CD33 Single Epitope CART In Vitro Cell Killing Experiment 1. Experimental objective and principle: Specific antibody clones screened from the phage library were constructed on lentiviral vectors with CAR structures, and their in vitro killing ability on corresponding CAR-T cells was studied in vitro by packaging and transducing human T cells with lentiviruses.

[0114] In vitro cell killing assays used THP-1 cells as CD33-positive target cells, and the antigen-specific killing ability of CD33 CAR-T cells was evaluated by comparing different combinations. The target cells were obtained through lentiviral transduction to stably express firefly luciferase; therefore, luciferase activity in the sample reflects the number of target cells. CAR-T cells and target cells were co-incubated. When target cells were killed by CAR-T cells, luciferase was released and quickly inactivated (firefly luciferase half-life is approximately 0.5 h). If target cells were not killed or inhibited by CAR-T cells, more luciferase would be produced as target cells proliferated and luciferase expression continued. Therefore, luciferase activity can be used to detect the CAR-T cell killing effect.

[0115] 2. Brief experimental steps for in vitro cell killing: 1) Centrifuge CD33-positive target cells (THP-1) at 500g for 5 min at room temperature, discard the supernatant, and resuspend in 1640+10% FBS medium to a concentration of 1x10⁻¹ 5 Cells / mL; Add 100 μL of target cells to each well of a 96-well plate; 2) Based on the CAR positivity rate and effector-to-target ratio of the CAR-T (VHH01, VHH19) samples to be tested, add 100uL of CAR-T cells to each well of a 96-well plate and mix them with the target cells; then incubate them in a carbon dioxide incubator. 3) Use a luciferase assay kit to detect the luciferase activity in each well sample.

[0116] Main samples and reagents: CAR T cells: VHH01, VHH19 CD33-positive target cells: THP-1-GFP-luci Stable-Lite® Luciferase Assay System, Novizan, DD1202.

[0117] Experimental results: like Figure 6 The VHH01 and VHH19CART cells shown can effectively kill CD33-positive THP-1 cells.

[0118] Example 6: CD33 dual-epitope CART in vitro cell killing experiment 1. Experimental objective and principle: Specific antibody clones screened from the phage library were constructed into lentiviral vectors with dual epitope CAR structures. The in vitro killing ability of the corresponding CAR-T cells was studied by packaging and transducing human T cells with lentiviruses.

[0119] In vitro cell killing assays used THP-1 and U937 cells as CD33-positive target cells, and the antigen-specific killing ability of CD33 CAR-T cells was evaluated by comparing different combinations. The target cells were obtained through lentiviral transduction to stably express firefly luciferase; therefore, luciferase activity in the sample reflects the number of target cells. CAR-T cells and target cells were co-incubated. When target cells were killed by CAR-T cells, luciferase was released and quickly inactivated (firefly luciferase half-life is approximately 0.5 h). If target cells were not killed or inhibited by CAR-T cells, more luciferase would be produced as target cells proliferated and luciferase expression continued. Therefore, luciferase activity can be used to detect the CAR-T cell killing effect.

[0120] 2. Brief experimental steps for in vitro cell killing: 1) Centrifuge the target cells at 500g for 5 min at room temperature, discard the supernatant, and resuspend them in 1640+10% FBS medium to a concentration of 1x10⁻¹. 5Cells / mL; Add 100 μL of target cells to each well of a 96-well plate; 2) Based on the CAR positivity rate and effector-to-target ratio of the CAR-T (VHH01-VHH19, VHH19-VHH01, GO) samples to be tested, add 100uL of CAR-T cells to each well of a 96-well plate and mix them with the target cells; then incubate them in a carbon dioxide incubator. 3) Use a luciferase assay kit to detect the luciferase activity in each well sample.

[0121] Main samples and reagents: CD33-positive target cells: THP-1-GFP-luci, U937-GFP-luci Stable-Lite® Luciferase Assay System, Novizan, DD1202.

[0122] Experimental results: such as Figure 7 The results show that VHH19-VHH01, VHH01-VHH19, and GO can effectively kill target cells when using THP-1 and U937 as target cells.

[0123] Example 7: In vitro cell killing experiment of CD33 dual-epitope CART blocking mTNF-α secretion 1. Experimental objective and principle: Specific antibody clones screened from the phage library were constructed into lentiviral vectors with dual epitopes and CAR structures overexpressing the TNF-α receptor, respectively. The in vitro killing ability of the corresponding CAR-T cells was studied by packaging and transducing human T cells with lentiviruses.

[0124] In vitro cell killing experiments used U937 cells as CD33-positive target cells, and the antigen-specific killing ability of CD33 CAR-T cells was evaluated by comparing different combinations. The target cells were obtained through lentiviral transduction to stably express firefly luciferase; therefore, luciferase activity in the sample reflects the number of target cells. CAR-T cells and target cells were co-incubated. When target cells were killed by CAR-T cells, luciferase was released and quickly inactivated (firefly luciferase half-life is approximately 0.5 h). If target cells were not killed or inhibited by CAR-T cells, more luciferase would be produced as target cells proliferated and luciferase expression continued. Therefore, luciferase activity can be used to detect the killing effect of CAR-T on target cells.

[0125] 2. Brief experimental steps for in vitro cell killing: 1) Centrifuge the target cells at 500g for 5 min at room temperature, discard the supernatant, and resuspend them in 1640+10% FBS medium to a concentration of 1x10⁻¹. 5 Cells / mL; Add 100 μL of target cells to each well of a 96-well plate; 2) Based on the CAR positivity rate and effector-target ratio of the CAR-T (VHH19-VHH01, VHH19-VHH01-TNF-αblock) samples to be tested, add 100uL of CAR-T cells to each well of a 96-well plate and mix them with the target cells; then incubate them in a carbon dioxide incubator. 3) Use a luciferase assay kit to detect the luciferase activity in each well sample.

[0126] Main samples and reagents: CD33-positive target cells: U937-GFP-luci Stable-Lite® Luciferase Assay System, Novizan, DD1202.

[0127] Experimental results: such as Figure 8 As shown, both VHH19-VHH01 and VHH19-VHH01-TNF-αblock can effectively kill CD33-positive target cells U937-GFP-luci in vitro.

[0128] Example 8: Detection of in vitro cell-killing cytokines 1) Centrifuge THP-1-GFP-luci at 500g for 5 min at room temperature, discard the supernatant, and resuspend in 1640+10% FBS medium to a concentration of 4x10⁻⁶. 5 Cells / mL; Add 250 μL of target cells to each well of a 24-well plate; 2) Based on the CAR positivity rate and effector-target ratio of 1:1 for the CAR-T samples VHH019-VHH01 and VHH19-VHH01-TNF-αblock, 250uL of CAR-T cells were added to each well of a 24-well plate and mixed with the target cells; then the plates were incubated in a CO2 incubator for 24 hours. 3) The secretion of TNF-α in each well sample was detected using a TNF-α elisa kit.

[0129] Experimental results: like Figure 9 As shown, the TNF-α secretion level of VHH19-VHH01-TNF-αblock was significantly lower than that of VHH19-VHH01.

[0130] Example 8: Antitumor efficacy experiment of CD33CAR-T in NCG tumor-bearing mice To further validate the antitumor efficacy of CD33 CAR-T cells in mice, the antitumor efficacy of VHH19-VHH01 and the positive control GO was first verified in mice. Then, the antitumor efficacy of VHH01, VHH19, VHH19-VHH01, and VHH19-VHH01-TNF-α block in mice was further verified. The tumor cell line U937-Luc, expressing firefly luciferase (ffLuc), was used to confirm its normal expression of firefly luciferase, and U937 itself is a naturally occurring cell line with high CD33 expression. U937-Luc cells were injected into immunodeficient NCG mice via tail vein. After tumor formation, candidate clonal CAR-T cells were inoculated to verify their antitumor effect.

[0131] Experimental procedure: A cell line U937-luc was established to overexpress firefly luciferase (ffLuc), and it was confirmed that it could express firefly luciferase normally. After the cells were cultured to the logarithmic growth phase, the cells were collected. Purchase NCG mice, feed them for 1 week, and inject them with U937-luc cells via tail vein, 2×10⁶ cells per mouse. 6 Cells / 200 μL PBS, and the general condition of mice was observed daily after injection; On day 7, mice were divided into groups and injected with CAR-T cells (5×10⁻⁶ cells) via the tail vein. 6 CAR+ cells / cell), control untransfected Mock-T cells.

[0132] Regular imaging observations, weight measurements, and general condition observations of mice were performed and recorded. T cell proliferation in mouse blood was also measured periodically.

[0133] Experimental results: 1) Verify the in vivo antitumor efficacy of VHH19-VHH01 and the positive control GO in mice, such as Figure 10 The figures show the survival curve, imaging luminescence intensity curve, and changes in CD3 levels in the blood, respectively. The tumor burden in the MOCK-T group continued to increase after CAR cell reinfusion. The tumor burden of VHH19-VHH01 and GO was significantly lower than that of MOCK-T, indicating significant tumor inhibition. Based on these results, VHH19-VHH01 effectively controlled the tumor burden in mice, and the positive control GO mice died before the VHH19-VHH01 experimental group. The changes in CD3 levels in the blood of the experimental groups indicate that both VHH19-VHH01 and the positive control GO ensured effective in vivo expansion, with VHH19-VHH01 showing the best results.

[0134] 2) Verify the in vivo antitumor efficacy of VHH01, VHH19, VHH19-VHH01, and VHH19-VHH01-TNF-α block in mice, such as... Figure 11 The figures show the survival curve, imaging luminescence intensity curve, and changes in CD3 levels in the blood, respectively. The tumor burden in the MCOK-T group continued to increase after CAR cell reinfusion. The tumor burden of VHH01, VHH19, VHH19-VHH01, and VHH19-VHH01-TNF-αblock was significantly lower than that of MOCK-T, indicating significant tumor inhibition. The changes in CD3 levels in the blood of the experimental groups show that VHH01, VHH19, VHH19-VHH01, and VHH19-VHH01-TNF-αblock can all ensure effective in vivo expansion.

[0135] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

[0136] Sequence List: The amino acid sequence of the VHH01 variable region is shown in SEQ ID NO: 1: QVQLVESGGGLVQPGGSLKLSCAASGDIGSLYSMGWYHEAPGKGREVVAVATNDGSTHYADSVRGRFTISRDNAKNTVDLQMNSLKPEDTAVYYCNAWAGPVGGRRFYWGQGTQVTVSS The nucleotide sequence of the VHH01 variable region is shown in SEQ ID NO: 2: CAGGTGGCAGCTGGTTGAATCTGGCGGCGGCTTGGTTCAGCCTGGCGGCAGCCTGAAGCTGTCCTGTGCCGCCAGTGGTGACATCGGCAGCCTCTACTCCATGGGCTGGTACCATGAGGCTCCTGGCAAGGGCAGAGAAGTGGTGGCTGTGGCCACCAATGATGGCTCTACCCACTACG CTGACTCCGTGCGGGGCCGCTTCACCATCTCCCGGGACAATGCCAAGAACACCGTGGACCTGCGATGAACTCCCTGAAGCCTGAGGACACCGCCGTGTACTACTGCAACGCCTGGGCTGGTCCTGTGGGTGGCCGGAGATTCTACTGGGGCCAGGGCACCCAGGTCACCGTGTCCTCT The VHH01 CDR1 sequence is shown in SEQ ID NO: 3: GDIGSLYS The VHH01 CDR2 sequence is shown in SEQ ID NO: 4: ATNDGST The VHH01 CDR3 sequence is shown in SEQ ID NO:5: NAWAGPVGGRRFY The amino acid sequence of the VHH19 variable region is shown in SEQ ID NO: 6: EVQLVESGGGLVQPGGSLRLSCAASGFTLDYYAIGWFRQAPGKEREGVLCISSSGGITDYTDSVKGRFTISRDNAKNTVYLQMNSLKPEDTAVYYCAGRTQRCYIDYDGHWSGYDYWGQGTQVTVSS The nucleotide sequence of the VHH19 variable region is shown in SEQ ID NO: 7: GAAGTGCAGCTGGTTGAATCTGGCGGCGGCTTGGTTCAGCCTGGAGGCTCTCTGAGACTGTCTTGTGCTGCCTCTGGCTTCACCCTGGACTACTACGCCATCGGCTGGTTCAGACAGGCCCCTGGCAAAGAAAGAGAAGGCGTGCTGTGCATCTCCAGCAGCGGCGGCATCACCGACTACACAGACTCCG TGAAGGGCAGATTCACCATCTCCCGGGACAACGCCAAGAACACCGTGTACCTGCAGATGAACTCCCTGAAGCCTGAGGACACCGCCGTGTACTACTGTGCCGGCAGAACCCAGAGATGCTACATCGACTATGATGGCCACTGGAGTGGCTATGACTATTGGGGCCAGGGCACCCAGGTCACCGTGTCCTCT The VHH19 CDR1 sequence is shown in SEQ ID NO: 8: GFTLDYYA The VHH19 CDR2 sequence is shown in SEQ ID NO: 9: ISSSGGIT The VHH19 CDR3 sequence is shown in SEQ ID NO:10: AGRTQRCYIDYDGHWSGYDY The amino acid sequence of VHH01 CAR is shown in SEQ ID NO: 11: MALPVTALLLPLALLLHAARPQVQLVESGGGLVQPGGSLKLSCAASGDIGSLYSMGWYHEAPGKGREVVAVATNDGSTHYADSVRGRFTISRDNAKNTVDLQMNSLKPEDTAVYYCNAWAGPVGGRRFYWGQGTQVTVSSTTTPAPRPPTPAPTIASQPLSLRPEACRPAAGGAVHTRGLDFACDIYIWAPLAGTCGVLLLSLVITLYCKRGRKKLLYIFKQPFMRPVQTTQEEDGCSCRFPEEEEGGCELRVKFSRSADAPAYKQGQNQLYNELNLGRREEYDVLDKRRGRDPEMGGKPRRKNPQEGLYNELQKDKMAEAYSEIGMKGERRRGKGHDGLYQGLSTATKDTYDALHMQALPPRAAAGSGEGRGSLLTCGDVEENPGPRMLLLVTSLLLCELPHPAFLLIPRKVCNGIGIGEFKDSLSINATNIKHFKNCTSISGDLHILPVAFRGDSFTHTPPLDPQELDILKTVKEITGFLLIQAWPENRTDLHAFENLEIIRGRTKQHGQFSLAVVSLNITSLGLRSLKEISDGDVIISGNKNLCYANTINWKKLFGTSGQKTKIISNRGENSCKATGQVCHALCSPEGCWGPEPRDCVSCRNVSRGRECVDKCNLLEGEPREFVENSECIQCHPECLPQAMNITCTGRGPDNCIQCAHYIDGPHCVKTCPAGVMGENNTLVWKYADAGHVCHLCHPNCTYGCTGPGLEGCPTNGPKIPSIATGMVGALLLLLVVALGIGLFM The nucleotide sequence of VHH01 CAR is shown in SEQ ID NO: 12: The amino acid sequence of VHH19 CAR is shown in SEQ ID NO: 13: MALPVTALLLPLALLLHAARPEVQLVESGGGLVQPGGSLRLSCAASGFTLDYYAIGWFRQAPGKEREGVLCISSSGGITDYTDSVKGRFTISRDNAKNTVYLQMNSLKPEDTAVYYCAGRTQRCYIDYDGHWSGYDYWGQGTQVTVSSTTTPAPRPPTPAPTIASQPLSLRPEACRPAAGGAVHTRGLDFACDIYIWAPLAGTCGVLLLSLVITLYCKRGRKKLLYIFKQPFMRPVQTTQEEDGCSCRFPEEEEGGCELRVKFSRSADAPAYKQGQNQLYNELNLGRREEYDVLDKRRGRDPEMGGKPRRKNPQEGLYNELQKDKMAEAYSEIGMKGERRRGKGHDGLYQGLSTATKDTYDALHMQALPPRAAAGSGEGRGSLLTCGDVEENPGPRMLLLVTSLLLCELPHPAFLLIPRKVCNGIGIGEFKDSLSINATNIKHFKNCTSISGDLHILPVAFRGDSFTHTPPLDPQELDILKTVKEITGFLLIQAWPENRTDLHAFENLEIIRGRTKQHGQFSLAVVSLNITSLGLRSLKEISDGDVIISGNKNLCYANTINWKKLFGTSGQKTKIISNRGENSCKATGQVCHALCSPEGCWGPEPRDCVSCRNVSRGRECVDKCNLLEGEPREFVENSECIQCHPECLPQAMNITCTGRGPDNCIQCAHYIDGPHCVKTCPAGVMGENNTLVWKYADAGHVCHLCHPNCTYGCTGPGLEGCPTNGPKIPSIATGMVGALLLLLVVALGIGLFM The nucleotide sequence of VHH19 CAR is shown in SEQ ID NO: 14: The amino acid sequence of VHH01-VHH19 CAR is shown in SEQ ID NO: 15: MALPVTALLLPLALLLHAARPQVQLVESGGGLVQPGGSLKLSCAASGDIGSLYSMGWYHEAPGKGREVVAVATNDGSTHYADSVRGRFTISRDNAKNTVDLQMNSLKPEDTAVYYCNAWAGPVGGRRFYWGQGTQVTVSSGGGGSGGGGSGGGGSEVQLVESGGGLVQPGGSLRLSCAASGFTLDYYAIGWFRQAPGKEREGVLCISSSGGITDYTDSVKGRFTISRDNAKNTVYLQMNSLKPEDTAVYYCAGRTQRCYIDYDGHWSGYDYWGQGTQVTVSSTTTPAPRPPTPAPTIASQPLSLRPEACRPAAGGAVHTRGLDFACDIYIWAPLAGTCGVLLLSLVITLYCKRGRKKLLYIFKQPFMRPVQTTQEEDGCSCRFPEEEEGGCELRVKFSRSADAPAYKQGQNQLYNELNLGRREEYDVLDKRRGRDPEMGGKPRRKNPQEGLYNELQKDKMAEAYSEIGMKGERRRGKGHDGLYQGLSTATKDTYDALHMQALPPRAAAGSGEGRGSLLTCGDVEENPGPRMLLLVTSLLLCELPHPAFLLIPRKVCNGIGIGEFKDSLSINATNIKHFKNCTSISGDLHILPVAFRGDSFTHTPPLDPQELDILKTVKEITGFLLIQAWPENRTDLHAFENLEIIRGRTKQHGQFSLAVVSLNITSLGLRSLKEISDGDVIISGNKNLCYANTINWKKLFGTSGQKTKIISNRGENSCKATGQVCHALCSPEGCWGPEPRDCVSCRNVSRGRECVDKCNLLEGEPREFVENSECIQCHPECLPQAMNITCTGRGPDNCIQCAHYIDGPHCVKTCPAGVMGENNTLVWKYADAGHVCHLCHPNCTYGCTGPGLEGCPTNGPKIPSIATGMVGALLLLLVVALGIGLFM The nucleotide sequence of VHH01-VHH19 CAR is shown in SEQ ID NO: 16: The amino acid sequence of VHH19-VHH01 CAR is shown in SEQ ID NO: 17: MALPVTALLLPLALLLHAARPEVQLVESGGGLVQPGGSLRLSCAASGFTLDYYAIGWFRQAPGKEREGVLCISSSGGITDYTDSVKGRFTISRDNAKNTVYLQMNSLKPEDTAVYYCAGRTQRCYIDYDGHWSGYDYWGQGTQVTVSSGGGGSGGGGSGGGGSQVQLVESGGGLVQPGGSLKLSCAASGDIGSLYSMGWYHEAPGKGREVVAVATNDGSTHYADSVRGRFTISRDNAKNTVDLQMNSLKPEDTAVYYCNAWAGPVGGRRFYWGQGTQVTVSSTTTPAPRPPTPAPTIASQPLSLRPEACRPAAGGAVHTRGLDFACDIYIWAPLAGTCGVLLLSLVITLYCKRGRKKLLYIFKQPFMRPVQTTQEEDGCSCRFPEEEEGGCELRVKFSRSADAPAYKQGQNQLYNELNLGRREEYDVLDKRRGRDPEMGGKPRRKNPQEGLYNELQKDKMAEAYSEIGMKGERRRGKGHDGLYQGLSTATKDTYDALHMQALPPRAAAGSGEGRGSLLTCGDVEENPGPRMLLLVTSLLLCELPHPAFLLIPRKVCNGIGIGEFKDSLSINATNIKHFKNCTSISGDLHILPVAFRGDSFTHTPPLDPQELDILKTVKEITGFLLIQAWPENRTDLHAFENLEIIRGRTKQHGQFSLAVVSLNITSLGLRSLKEISDGDVIISGNKNLCYANTINWKKLFGTSGQKTKIISNRGENSCKATGQVCHALCSPEGCWGPEPRDCVSCRNVSRGRECVDKCNLLEGEPREFVENSECIQCHPECLPQAMNITCTGRGPDNCIQCAHYIDGPHCVKTCPAGVMGENNTLVWKYADAGHVCHLCHPNCTYGCTGPGLEGCPTNGPKIPSIATGMVGALLLLLVVALGIGLFM The nucleotide sequence of VHH19-VHH01 CAR is shown in SEQ ID NO: 18: The amino acid sequence of VHH01-VHH19-TNFαblock CAR is shown in SEQ ID NO: 19: MALPVTALLLPLALLLHAARPQVQLVESGGGLVQPGGSLKLSCAASGDIGSLYSMGWYHEAPGKGREVVAVATNDGSTHYADSVRGRFTISRDNAKNTVDLQMNSLKPEDTAVYYCNAWAGPVGGRRFYWGQGTQVTVSSGGGGSGGGGSGGGGSEVQLVESGGGLVQPGGSLRLSCAASGFTLDYYAIGWFRQAPGKEREGVLCISSSGGITDYTDSVKGRFTISRDNAKNTVYLQMNSLKPEDTAVYYCAGRTQRCYIDYDGHWSGYDYWGQGTQVTVSSTTTPAPRPPTPAPTIASQPLSLRPEACRPAAGGAVHTRGLDFACDIYIWAPLAGTCGVLLLSLVITLYCKRGRKKLLYIFKQPFMRPVQTTQEEDGCSCRFPEEEEGGCELRVKFSRSADAPAYKQGQNQLYNELNLGRREEYDVLDKRRGRDPEMGGKPRRKNPQEGLYNELQKDKMAEAYSEIGMKGERRRGKGHDGLYQGLSTATKDTYDALHMQALPPRAAAGSGEGRGSLLTCGDVEENPGPMALPVTALLLPLALLLHAARPEVQLVESGGGLVQPGRSLRLSCAASGFTFDDYAMHWVRQAPGKGLEWVSAITWNSGHIDYADSVEGRFTISRDNAKNSLYLQMNSLRAEDTAVYYCAKVSYLSTASSLDYWGQGTLVTVSSGSTSGSGKPGSGEGSTKGDIQMTQSPSSLSASVGDRVTITCRASQGIRNYLAWYQQKPGKAPKLLIYAASTLQSGVPSRFSGSGSGTDFTLTISSLQPEDVATYYCQRYNRAPYTFGQGTKVEIKTTTPAPRPPTPAPTIASQPLSLRPEACRPAAGGAVHTRGLDFACDIYIWAPLAGTCGVLLLSLVITLYC The nucleotide sequence of VHH01-VHH19-TNFαblock CAR is shown in SEQ ID NO: 20:

Claims

1. An anti-CD33 nanobody or an antigen-binding fragment thereof, wherein the nanobody or the antigen-binding fragment thereof is selected from at least one of the following (1)-(4): (1) The amino acid sequences of the nanobody or its antigen-binding fragment CDR1, CDR2 and CDR3 are the three CDRs of the sequence shown in SEQ ID NO:1, or have 1, 2 or 3 amino acid substitutions, deletions or additions compared with the three CDRs of the sequence shown in SEQ ID NO:1; preferably, they have 1 conserved amino acid substitution. (2) The amino acid sequences of the nanobody or its antigen-binding fragment CDR1, CDR2 and CDR3 are the three CDRs of the sequence shown in SEQ ID NO:6, or have 1, 2 or 3 amino acid substitutions, deletions or additions compared with the three CDRs of the sequence shown in SEQ ID NO:6; preferably, they have 1 conserved amino acid substitution. (3) The amino acid sequences of CDR1, CDR2, and CDR3 of the nanobody or its antigen-binding fragment are as shown in SEQ ID NO:3, SEQ ID NO:4, and SEQ ID NO:5, respectively, or have 1, 2, or 3 amino acid substitutions, deletions, or additions compared to SEQ ID NO:3, SEQ ID NO:4, and SEQ ID NO:5; preferably, they have 1 conserved amino acid substitution. (4) The amino acid sequences of the nanobody or its antigen-binding fragment CDR1, CDR2, and CDR3 are as shown in SEQ ID NO:8, SEQ ID NO:9, and SEQ ID NO:10, respectively, or have 1, 2, or 3 amino acid substitutions, deletions, or additions compared to SEQ ID NO:8, SEQ ID NO:9, and SEQ ID NO:10; preferably, they have 1 conserved amino acid substitution.

2. The nanobody or its antigen-binding fragment according to claim 1, characterized in that, The amino acid sequence of the nanobody or its antigen-binding fragment is as shown in SEQ ID NO:1, or an amino acid sequence having at least 75%, at least 80%, at least 85%, at least 90%, at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% identity with SEQ ID NO:1; or the amino acid sequence of the nanobody or its antigen-binding fragment is as shown in SEQ ID NO:6, or an amino acid sequence having at least 75%, at least 80%, at least 85%, at least 90%, at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% identity with SEQ ID NO:6; Preferably, the nanobody or its antigen-binding fragment specifically binds to the IgV domain or IgC2 domain of CD33; more preferably, the nanobody shown in (1) or (3) or its antigen-binding fragment specifically binds to the IgV domain of CD33; the nanobody shown in (2) or (4) or its antigen-binding fragment specifically binds to the IgC2 domain of CD33.

3. The nanobody or its antigen-binding fragment according to claim 1, characterized in that, The nanobody or its antigen-binding fragment can be linked, fused, or combined with one or more nanobodies, antibodies, or their antigen-binding fragments with different binding specificities to form bispecific, trispecific, or multispecific antibody molecules or compositions; the different binding specificities include binding specificities against different antigens, different antigenic epitopes, or different functional targets. A preferred method is to form a bispecific antibody molecule by combining a nanobody or its antigen-binding fragment that specifically binds to the CD33 IgV domain (preferably the nanobody or its antigen-binding fragment shown in (1) or (3)) with a nanobody or its antigen-binding fragment that specifically binds to the CD33 IgC2 domain (preferably the nanobody or its antigen-binding fragment shown in (2) or (4)). The preferred method is to form a trispecific antibody molecule by specifically binding a nanobody or its antigen-binding fragment to the CD33 IgV domain (preferably the nanobody or its antigen-binding fragment shown in (1) or (3)), specifically binding a nanobody or its antigen-binding fragment to the CD33 IgC2 domain (preferably the nanobody or its antigen-binding fragment shown in (2) or (4)), and a TNF-α binding molecule (preferably an anti-TNF-α antibody, its antigen-binding fragment, or an anti-TNF-α nanobody).

4. A chimeric antigen receptor (CAR) comprising one or more antigen-binding domains that specifically bind to CD33, said antigen-binding domain comprising the nanobody of claim 1 or an antigen-binding fragment thereof.

5. A chimeric antigen receptor (CAR) according to claim 4, characterized in that, The CAR further includes at least one of the following: (b) Hinge region, preferably the hinge region includes the hinge region of the following molecules or variants thereof: CD8α, CD28, IgG1, IgG4, CD34; more preferably the CD8α hinge region, the IgG4 hinge region or variants thereof; (c) Transmembrane region, preferably comprising transmembrane domains of the following molecules or variants thereof: CD8α, CD28, CD3ζ, ICOS, 4-1BB (CD137), OX40 (CD134), PD-1, CD4, TCRα, TCRβ, CD3ε, CD3δ, CD3γ; more preferably CD8α transmembrane region, CD28 transmembrane region or variants thereof; (d) An intracellular signal transduction domain comprising at least one co-stimulatory domain and / or one intracellular signal transduction domain, wherein the co-stimulatory domain preferably comprises a co-stimulatory signal transduction domain of the following molecules or variants thereof: 4-1BB (CD137), CD28, OX40, ICOS (CD278), CD27, GITR (CD357), CD30, HVEM, DR3, CD226 (DNAM-1), 2B4 (CD244), DAP10, DAP12; more preferably 4-1BB, CD28 or a combination thereof; the intracellular signal transduction domain preferably comprises an intracellular signal transduction domain of the following molecules or variants thereof: CD3ζ, FcRγ, DAP10, DAP12, CD3ε, CD3γ, CD3δ; more preferably CD3ζ; Preferably, the CAR further comprises a signal peptide sequence, wherein the signal peptide is preferably an Igκ chain signal peptide, an immunoglobulin heavy chain signal peptide, a CD8α signal peptide, a CD4 signal peptide, a CD5 signal peptide, a CD28 signal peptide, a T cell receptor α chain (TCRα) signal peptide, a T cell receptor β chain (TCRβ) signal peptide, a GM-CSF signal peptide, or a functional variant thereof; preferably, it is a CD8α signal peptide, an Igκ chain signal peptide, or a functional variant thereof. Preferably, the CAR further comprises a self-cleaving peptide selected from T2A, P2A, E2A, F2A or functional variants thereof, preferably T2A or P2A; Preferably, the CAR further includes a detection tag and / or an auxiliary functional element; preferably, the detection tag may be selected from truncated epidermal growth factor receptor (tEGFR), truncated CD34 (tCD34), truncated CD19 (tCD19), truncated CD20 (tCD20), truncated CD22 (tCD22), RQR8, or a combination thereof, preferably tEGFR; preferably, the auxiliary functional element may be selected from immune checkpoint or immune regulatory target binding molecules, cytokines, cytokine receptors, or functional fragments thereof; wherein, the immune detection The point or immunomodulatory target binding molecules include antibodies, antibody fragments, or nanobodies targeting PD-1, PD-L1, CTLA-4, LAG-3, TIM-3, TIGIT, CD226, CD155, CD47, B7-H3, B7-H4, etc.; the cytokines or cytokine receptors include IL-2, IL-2 receptor, IL-7, IL-7 receptor, IL-15, IL-15 receptor, IL-12, IL-18, IL-21, CCL19, CCL21, or functional variants thereof.

6. A chimeric antigen receptor (CAR) according to claim 4, characterized in that, The CAR is a single-target CAR, whose antigen-binding domain specifically binds to CD33; Preferably, the single-target chimeric antigen receptor is a single epitope CAR; preferably, it includes, in sequence, an optional signal peptide, an antigen-binding domain that specifically binds to CD33, a hinge region, a transmembrane domain, a co-stimulatory signaling domain, and an intracellular signal transduction domain; preferably, the antigen-binding domain that specifically binds to CD33 is any one of (1)-(4) nanobodies or their antigen-binding fragments; the hinge region is a CD8α hinge region, the transmembrane region is a CD8α transmembrane domain, the co-stimulatory signaling domain is a 4-1BB co-stimulatory signaling domain, and the intracellular signal transduction domain is a CD3ζ intracellular signal transduction domain; Preferably, the single-target chimeric antigen receptor is a dual-epitope CAR, wherein the antigen-binding domain includes: (a1) First antigen-binding domain, which specifically binds to the CD33 IgV domain; (a2) The second antigen-binding domain, which specifically binds to the CD33 IgC2 domain; Preferably, the dual-epitope CAR comprises, sequentially connected, an optional signal peptide, a first antigen-binding domain, a linker peptide, a second antigen-binding domain, a hinge region, a transmembrane domain, a co-stimulatory signaling domain, and an intracellular signal transduction domain; preferably, the first antigen-binding domain is the nanobody or its antigen-binding fragment described in (1) or (3); the second antigen-binding domain is the nanobody or its antigen-binding fragment described in (2) or (4); the hinge region is the CD8α hinge region, the transmembrane region is the CD8α transmembrane domain, the co-stimulatory signaling domain is the 4-1BB co-stimulatory signaling domain, and the intracellular signal transduction domain is the CD3ζ intracellular signal transduction domain; Preferably, the CAR is a multi-target chimeric antigen receptor, comprising the single-target chimeric antigen receptor and a third antigen-binding domain, wherein the third antigen-binding domain specifically binds to immunosuppressive factors, immune checkpoint molecules, inflammatory factors, or tumor microenvironment regulators; preferably, the third antigen-binding domain specifically binds to TNF-α, PD-1, PD-L1, CTLA-4, LAG-3, TIM-3, TIGIT, CD47, B7-H3, B7-H4, or combinations thereof; more preferably, the third antigen-binding domain specifically binds to TNF-α; Preferably, the multi-target chimeric antigen receptor comprises, in sequence, an optional signal peptide, a first antigen-binding domain, a linker peptide, a second antigen-binding domain, a hinge region, a transmembrane domain, a co-stimulatory signaling domain, an intracellular signal transduction domain, a self-cleaving peptide, and a third antigen-binding domain; preferably, the first antigen-binding domain is the nanobody or its antigen-binding fragment described in (1) or (3); the second antigen-binding domain is the nanobody or its antigen-binding fragment described in (2) or (4); the third antigen-binding domain is an anti-TNF-α nanobody; the hinge region is a CD8α hinge region, the transmembrane region is a CD8α transmembrane domain, the co-stimulatory signaling domain is a 4-1BB co-stimulatory signaling domain, and the intracellular signal transduction domain is a CD3ζ intracellular signal transduction domain.

7. A nucleic acid molecule encoding the nanobody of claim 1 or its antigen-binding fragment, or the chimeric antigen receptor of claim 4.

8. An expression vector comprising the nucleic acid molecule of claim 7.

9. A host cell comprising the expression vector of claim 8.

10. An engineered immune cell expressing the chimeric antigen receptor of claim 4; Preferably, the immune cells include, but are not limited to, αβ T cells, γδ T cells, NK cells, NKT cells, cytokine-induced killer cells (CIK), macrophages, and any combination thereof; more preferably, CAR-T cells; Preferably, the immune cells can be autologous, allogeneic, or derived from induced pluripotent stem cell (iPSC) differentiation.

11. A pharmaceutical composition comprising the nanobody or antigen-binding fragment of claim 1, the chimeric antigen receptor of claim 4, the nucleic acid molecule of claim 7, the expression vector of claim 8 or the host cell of claim 9, the immune cell (preferably CAR-T cell) of claim 10, or a combination thereof, and a pharmaceutically acceptable carrier. Preferably, the pharmaceutical composition may further include other therapeutic agents, such as immune checkpoint inhibitors, cytokines, demethylating agents, etc., or be administered in combination with TNF-α blockers in the treatment regimen to enhance the anti-tumor synergistic effect.

12. Use of the nanobody or antigen-binding fragment of claim 1, the chimeric antigen receptor of claim 4, the nucleic acid molecule of claim 7, the expression vector of claim 8, or the host cell of claim 9, the immune cell of claim 10, or the pharmaceutical composition of claim 11 in the preparation of a medicament for treating diseases expressing CD33; The disease is preferably leukemia, and more preferably acute myeloid leukemia (AML), myelodysplastic syndrome (MDS), chronic myelomonocytic leukemia (CMML), juvenile myelomonocytic leukemia (JMML) and other myeloid malignant tumors that express CD33. Preferably, the drug can also be used to treat patients who have received myeloablative or non-myeloablative pretreatment, and to bridge hematopoietic stem cell transplantation (HSCT) after treatment to achieve long-term hematopoietic reconstitution and disease remission.

Images