WO2024129988A1

WO2024129988A1 - Compositions and methods for delivery of therapeutic agents to bone

Info

Publication number: WO2024129988A1
Application number: PCT/US2023/084054
Authority: WO
Inventors: Adrienne Marie ROTHSCHILDS; Nicholas McCartney PLUGIS; Charlotte Marie NICOD; Stephen Marshall; Avak Kahvejian; Yann Paul Guy Regis ECHELARD; Noubar Boghos Afeyan; Raffi AFEYAN; Scott Moore CARLSON; Vivek KOHAR; Francisco Javier CULLERE LUENGO; Andrew Kinloch; Nabil AZHAR; Daniel BLOM; Tao FANG
Original assignee: Flagship Pioneering Innovations VII Inc
Current assignee: Flagship Pioneering Innovations VII Inc
Priority date: 2022-12-14
Filing date: 2023-12-14
Publication date: 2024-06-20
Anticipated expiration: 2025-06-14
Also published as: WO2024129988A8; TW202430215A

Abstract

Macromolecule compositions and related methods that effect targeted delivery of therapeutic agents to effector targets in bone tissue while minimizing or avoiding undesirable delivery to other cells, tissues or organs are provided. Compositions and methods related to macromolecules, such as an ANDbody™, that include an effector target binding domain specific for an effector target, and an address binding domain specific for bone tissue are described.

Description

COMPOSITIONS AND METHODS FOR DELIVERY OF THERAPEUTIC AGENTS TO BONE

BACKGROUND OF THE INVENTION

Undesirable off-target effects are a problem for otherwise desirable therapeutic targets that are present in healthy as well as diseased tissues.

SUMMARY OF THE INVENTION

The present disclosure describes, in part, macromolecule compositions and related methods that effect targeted delivery of therapeutic agents to effector targets in a bone cell, tissue, or organ while minimizing or avoiding undesirable delivery to other cells, tissues or organs. Generally, compositions described herein comprise macromolecules, such as an ANDbody™, that include an effector target binding domain specific for bone, and an address binding domain specific for an address target. The address target is generally sufficiently restricted in the subject to target the macromolecule to the desired bone cell, tissue or organ. In some embodiments, the effector target binding domain does not influence an effector target in the absence of an address target binding domain. Moreover, the address target binding domain does not influence signaling upon binding the address target. However, localization of the effector target binding domain by the address target binding domain enables the effector target binding domain to bind the effector target sufficiently to elicit an influence on signaling by the effector target in the target cell or tissue. In addition, the macromolecules described herein may be linked to one or more small molecules. The compositions described herein can be used, e.g., to specifically deliver a therapeutic agent (for example, the effector target binding domain, the small molecule, or both) to a desired location, e.g., a bone cell, tissue or organ, in a subject, while avoiding undesirable off-target effects (e.g., undesirable off-target effects in the brain; skin; cardiovascular system, e.g., heart or vasculature) and/or avoiding certain toxicity, e.g., avoiding cardiovascular disease, such as stroke, myocardial infarction.

In one aspect, the present disclosure provides a method of localizing a macromolecule at a bone tissue or bone cell of a subject, the method comprising administering to the subject a macromolecule comprising a first binding site and a second binding site, wherein (a) the first binding site is specific for an effector target in a bone tissue or bone cell in the subject, and (b) the second binding site is specific for an address target expressed in a bone tissue or bone cell in the subject; wherein (i) the second binding site localizes the first binding site to the address target such that the first binding site influences effector target signaling in the bone tissue or bone cell; (ii) the second binding site does not substantially influence signaling upon binding the address target; and (iii) the first binding site does not substantially influence effector target signaling in the absence of localization by the second binding site; and allowing the macromolecule to localize at the bone tissue or bone cell of the subject.

In some embodiments, at least 25% of the macromolecule detectable in the subject is detected at the bone tissue or bone cell at a time point between 1 and 7 days following administration of the macromolecule to the subject. In some embodiments, the potency of the first binding site at the bone tissue or bone cell is substantially increased relative to a reference macromolecule lacking the second binding site.

In some embodiments, the first binding site has a low affinity for the effector target.

In some embodiments, the first binding site has a low avidity for the effector target.

In some embodiments, the affinity of the first binding site for the effector target is lower than the affinity of the second binding site for the address target.

In some embodiments, the avidity of the first binding site for the effector target is lower than the avidity of the second binding site for the address target.

In some embodiments, effector target signaling by the macromolecule in a non-target tissue or cell of the subject is substantially decreased relative to a reference macromolecule lacking the second binding site.

In some embodiments, the address target is regionally expressed in the subject. In some embodiments, the address target is locally expressed in the subject. In some embodiments, the expression of the address target is restricted to a cell type in the subject.

In some embodiments, the address target is expressed only by a cell in the subject when in a specific cell state. In some embodiments, the address target is expressed only by a cell in the subject in a disease state.

In some embodiments, the first binding site or the second binding site comprises a polypeptide. In some embodiments, the polypeptide is an antibody or antigen-binding fragment thereof. In some embodiments, the macromolecule is an antibody comprising a first binding site that is specific for the effector target in the subject and a second binding site that is specific for the address target.

In some embodiments, the polypeptide is a ligand of the effector target or a ligand of the address target. In some embodiments, (a) the first binding site comprises an antibody or antigenbinding fragment thereof and the second binding site comprises a ligand of the address target; or (b) the first binding site comprises a ligand of the effector target and the second binding site comprises an antibody or antigen-binding fragment thereof.

In some embodiments, (a) the first binding site comprises an antibody or antigen-binding fragment thereof and the second binding site comprises a small molecule that binds to the address target; or (b) the first binding site comprises a small molecule that binds to the effector target and the second binding site comprises an antibody or antigen-binding fragment thereof.

In some embodiments, the second binding site is specific for chondroadherin (CHAD). In some embodiments, the second binding site is specific for dentin matrix acidic phosphoprotein 1 (DMP1 ). In some embodiments, the second binding site is specific for integrin binding sialoprotein (IBSP). In some embodiments, the second binding site is specific for trophoblast glycoprotein (TPBG). In some embodiments, the second binding site is specific for hydroxyapatite. In some embodiments, the second binding site is a bisphosphonate (e.g., alendronate, risenodrate, etidronate, ibandronate, clodronate, tiludronate, pamidronate, or zoledronate). In some embodiments, the first binding site is specific for sclerostin (SOST). In some embodiments, the first binding site is specific for dickkopf-1 (DKK1 ).

In some embodiments, the second binding site is specific for interferon-induced transmembrane protein 5 (IFITM5). In some embodiments, the first binding site is specific for SOST.

In some embodiments, the macromolecule is linked to a small molecule. In some embodiments, the small molecule is odanacatib. In some embodiments, the small molecule is a bisphosphonate (e.g., alendronate, risenodrate, etidronate, ibandronate, clodronate, tiludronate, pamidronate, or zoledronate).

In some embodiments, the macromolecule and the small molecule are connected by a linker. In some embodiments, the linker is a cleavable linker. In some embodiments, the linker is a non- cleavable linker.

The details of one or more embodiments of the invention are set forth in the description below. Other features or advantages of the present invention will be apparent from the following drawings and detailed description of several embodiments, and also from the appended claims.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 is a schematic illustrating exemplary ANDbody™ molecules and their use as logicgated medicines. FIG. 1 shows broad distribution of a therapeutic target (right side), such as an effector target, in a human subject with no address targeting, and a localized and restricted distribution with address targeting (left side), which is provided by an address target binding domain. FIG 1 . also provides a representative bipartite structure of an ANDbody with an address target binding domain linked to an effector target binding domain, which includes a functional moiety, e.g., a moiety that modulates, e.g., agonizes or antagonizes, a target effector in an address targeted cell or tissue. The address target binding domain directs the ANDbody to a desired location, such as a targeted cell or tissue, allowing for the effector target binding domain to engage the therapeutic effector target in the localized and restricted distribution area. In some embodiments, high affinity of the effector domain for the target effector may not be required; localization of the effector target binding domain by the address target binding domain enables the effector target binding domain to bind the effector target sufficiently to elicit an influence on signaling by the effector target in the target cell or tissue despite low affinity of the effector domain for the effector target. The address target binding domain can alternatively be used to transport molecular or cellular cargos to a desired address.

FIG. 2 is a schematic map showing activity of exemplary effector targets that can be restricted to tissues or cells of interest by developing ANDbody therapeutics comprised of an effector targeting domain and an address targeting domain. These ANDbody biologies represent potent, address- restricted medicines according to the present technology.

FIG. 3 provides exemplary structures of ANDbody biologies that can be engineered according to the present technology, including (but not limited to): an asymmetric antibody, an dual-affinity retargeting protein (DART), a tandem diabody (TandAb), a diabody, an Fab2, lgG(L,H)-Fv or a BiTE. FIG. 4 demonstrates an EC50 curve of an exemplary single effector targeting domain (dashed line), such as a monospecific biologic (for example scFv) having a single binding domain to an effector target compared to an EC50 of an exemplary bispecific ANDbody biologic (for example di- scFc) with an address target binding domain and an effector target binding domain (solid line), such that single effector targeting domain (usually broadly expressed) is targeted/restricted to local, address target-specific tissues and/or cells, thus effectively increasing affinity of the effector target binding domain for the effector target binding site, as evidenced by a shift of the curve to the left (lower EC50, higher affinity).

FIG. 5 is a set of representative images showing the longitudinal interface of bone marrow and the trabecular region (femur ‘cap’) in mouse femur sections stained using an anti-dentin matrix acidic phosphoprotein 1 monoclonal antibody (anti-DMP1 mAb) (right panel) or incubated with only the horseradish peroxidase (HRP) secondary (2°) reagent (control; left panel). The HRP 2° reagent was from the Mouse on Mouse Polymer IHC Kit (Abeam ab269452).

FIG. 6 is a graph showing levels of an anti-TPBG monoclonal antibody (Ab) detected on HEK293T cells that were transiently transfected to express full-length mouse or human TPBG constructs tagged at the C-terminus with intracellular FLAG peptide (C FLAG). The cells were incubated with a serial dilution of the anti-TPBG monoclonal antibody, followed by detection with a fluorophore-conjugated secondary antibody. Levels are expressed as mean fluorescence intensity (MFI).

FIG. 7 is a set of representative micrographs showing binding of an anti-TPBG mAb (left panel) to MC3T3 osteoblastic cells grown on 96-well plates, as indicated by fluorescent signal from a fluorophore secondary antibody. Binding of an equivalent concentration of a control murine IgG is shown as a control (right panel).

Fig. 8 is a diagram showing the structure of BIS-L1 .

FIG. 9 is a graph showing the results of a Wnt1 luciferase reporter assay (measured as relative light units (RLU) of luciferase) in HEK293 reporter cells treated with the indicated amounts of an anti-SOST antibody (PRO136) or a bisphosphonate-conjugated anti-SOST ANDbody (anti-SOST- BIS-L1 (2)).

FIG. 10 is a graph showing the proportion of anti-SOST-BIS-L1 (2) or the unconjugated anti- SOST ANDBody PRO136 that was bound to hydroxyapatite (HA) in an in vitro assay at the indicated time points.

FIG. 1 1 A is an image of a decalcified mouse femur section that was collected eight days after subcutaneous administration of a bisphosphonate-conjugated anti-SOST ANDbody (anti-SOST-BIS- L1 (2)). The ANDbody was detected using an alkaline phosphatase-conjugated anti-human IgG antibody, which was developed using VECTOR® Red substrate.

FIG. 1 1 B is an image of a decalcified mouse femur section that was collected eight days after subcutaneous administration of 20 milligrams per kilogram of body weight (mpk) an anti-SOST ANDbody (anti-SOST - PRO136). The ANDbody was detected using an alkaline phosphatase- conjugated anti-human IgG antibody, which was developed using VECTOR® Red substrate. FIG. 12 is a graph showing levels of procollagen type 1 N-terminal propeptide (P1 NP) in serum from mice at the indicated timepoints following a single subcutaneous injection of 20 mpk of a bisphosphonate-conjugated anti-SOST ANDbody (anti-SOST-BIS-L1 (2)), an anti-SOST antibody (anti-SOST - PRO136) (positive control), an anti-RSV antibody (PRO022) (negative control), or a vehicle control (PBS). D8: Day 8.

DETAILED DESCRIPTION OF THE INVENTION

Provided herein are ANDbody™ molecules that include a therapeutic effector target binding domain and an address target binding domain, wherein the address target is a bone tissue or cell. In some embodiments, the ANDbody molecule is linked to a small molecule or more than one small molecule. The therapeutic effector target on the ANDbody molecule productively engages its therapeutic effector target only if the address target binding domain also engages an address target on a target tissue or cell to localize the effector target to the targeted cell or tissue, e.g., to form an AND-gate type of logic gate. For example, in some embodiments, an ANDbody is a macromolecule comprising at least (a) a first binding site specific for a therapeutic effector target that is expressed, e.g., broadly expressed, on a mammalian subject, e.g., on a cell surface; and (b) a second binding site specific for an address target, wherein the address target is a bone tissue or cell. In embodiments, expression of the address target is restricted to a bone tissue or cell in vivo in a subject. In some embodiments, the binding of a first binding site to a therapeutic effector target is weaker than the binding of the second binding site to the address marker. The effector and address targets may be on the same cell, or in different cells or compartments within the same tissue.

In some embodiments, at least 25% of the macromolecule (e.g., ANDbody) detectable in the subject is detected at the target bone tissue or cell at a time point between 1 and 7 days (e.g., at 1 day, 2 days, 3 days, 4 days, 5, days, 6 days, and/or 7 days) following administration of the macromolecule (e.g., ANDbody) to the subject. For example, in some embodiments, at least 30%, at least 35%, at least 40%, at least 45%, at least 50%, at least 55%, at least 60%, at least 65%, at least 70%, at least 75%, at least 80%, at least 85%, at least 90%, at least 95%, at least 98%, or 100% (e.g., 25-30%, 30-35%, 35-40%, 40-45%, 45-50%, 50-55%, 55-60%, 60-65%, 65-70%, 70-75%, 75- 80%, 80-85%, 85-90%, 90-95%, or 95-100%) of the macromolecule detectable in the subject is detected at the target bone tissue or cell at a time point between 1 and 7 days following administration of the macromolecule the subject.

Bone Effector Target

An ANDbody™ of the invention comprises an effector that modulates a therapeutic effector target in a subject, e.g., a mammalian subject such as a human, in need thereof. As used herein, an "effector target" is a discrete structure (e.g., a cell surface protein, a transmembrane protein, a receptor) of a cell or tissue of a subject, to which a therapeutic effector binding domain of an ANDbody can bind and exert a modulating effect, such as a therapeutic effect, on the subject. The ANDbody described herein has a binding site specific for an effector target. Upon binding of the effector binding domain to the effector target, the effector modulates the target bone cell or tissue to produce a biological response, such as a therapeutic effect, on the subject. However, in some embodiments, the effector target binding domain provided herein may not elicit a biological effect unless it is provided in conjunction with an address targeting domain to localize the effector to the desired target address in a targeted bone cell or tissue. In some embodiments, such therapeutic signaling may require the binding of multiple effector targets by multiple macromolecules according to the invention.

In some embodiments, an effector target binding domain may produce a small/weak biological effect when provided alone and provide a larger/stronger biological effect when provided in conjunction with an address targeting domain that localizes and concentrates/focuses the effector to the desired target address in a targeted bone cell or tissue. In some embodiments, an effector target binding domain may produce an acceptable biological effect when provided alone and provide an even larger/stronger biological effect when provided in conjunction with an address targeting domain to localize the effector target binding domain to a targeted bone cell or tissue. In some embodiments, an effector target binding domain may produce a strong biological effect when provided alone and provide a strong, or stronger, targeted effect when provided in conjunction with an address targeting domain to localize the effector target binding domain to a targeted bone cell or tissue. In some embodiments, an effector target binding domain may produce a biological effect with undesirable off target biological effects when provided alone, but can be targeted, concentrated, and focused to desired addresses in a targeted bone cell or tissue when provided in conjunction with an address targeting domain in order to decrease or eliminate undesirable off-target biological effects. Accordingly, effector target binding domains of the present technology provide superior therapeutic agents that provide stronger, targeted biological effects with less side effects, including less unintended off-target biological effects, when provided in conjunction with address target binding domains as described herein.

Examples of such therapeutic signaling effects include, but are not limited to:

(i) blocking a signal transduction pathway that promotes or maintains a disease state;

(ii) activating a signal transduction pathway that reduces or prevents a disease state;

(iii) promoting antibody-dependent cellular cytotoxicity (ADCC);

(iv) inducing complement activation on the target cell or tissue;

(v) promoting phagocytosis;

(vi) blocking or activating a signal transduction pathway that promotes differentiation of a cell;

(vii) inducing tissue remodeling to reduce or prevent fibrosis.

In some embodiments, the therapeutic effector target is more broadly expressed than the address target in the subject. In some embodiments, the therapeutic effector target is expressed systemically, regionally, or locally in the organism. “Systemic expression” of a therapeutic effector target means that the therapeutic effector target is expressed at substantially the same levels in most parts of a subject organism body. Systemic expression involves a plurality of tissues. “Regional expression” of a therapeutic effector target means that the therapeutic target is expressed in an area less than systemic expression but more than local expression. Regional expression is not limited to a single tissue but can occur in a plurality of different tissues. “Local expression” of a therapeutic effector target means that the therapeutic target is expressed in single or few tissue areas. Local expression is not limited to a single tissue but can occur in a plurality of different tissues.

In some embodiments, the effector target binding domain has a low affinity for the effector target. For example, a low affinity may be an affinity of greater than 10 nM (e.g., an affinity between 10 nM-1 pM, e.g., an affinity between 10 nM and 100 nM). In some aspects, the effector target binding domain has an affinity (KD) for the effector target of equal to or less than 100 nM as measured using biolayer interferometry

In some embodiments, the effector target binding domain has a low avidity for the effector target. Non-limiting examples of therapeutic effector targets that can be targeted with ANDbodies disclosed herein are listed in Table 1 , along with the exemplary function for the effector targets.

Table 1 : Exemplary Effector Targets

In some embodiments, the effector target is sclerostin (SOST) and the effector target binding domain comprises a SOST inhibitor (e.g., romosozumab, blosozumab, setrusumab, SHR-1222, or a sclerostin inhibitor provided in Yu et al., Acta Pharm Sin B, 12(5): 2150-2170, 2022 or an antigenbinding fragment thereof).

In some embodiments, the SOST inhibitor is an anti-SOST antibody comprising the three heavy chain CDRs given by SEQ ID NOs: 245, 246, and 247 of U.S. Patent No. 8,017,120 B2 and the three light chain CDRs given by SEQ ID NOs.: 78, 79, and 80 of U.S. Patent No. 8,017,120 B2, which six CDRs are incorporated by reference. In some embodiments, the anti-SOST antibody is romosozumab. In some embodiments, the anti-SOST antibody is a variant of the foregoing, such as a variant of romosozumab, e.g., in some embodiments, is a variant comprising at least: 1 , 2, 3, 4, 5, 6, 7, 8, 9, or 10 amino acid substitutions relative to romosozumab, including non-conservative, conservative, or highly-conservative substitutions (or any combination), optionally where the one or more substitutions are conservative substitutions in the CDRs, e.g., in certain embodiments, in CDRs other than hCDR3 and/or other than ICDR1 . In certain embodiments, an anti-SOST antibody useful in the invention comprises a substitution to facilitate conjugation (e.g., comprises such a substitution relative to romosozumab, relative to an antibody comprising the 6 CDRs of romosozumab, or relative to a variant of any of the foregoing), such as a S239C substitution.

In some embodiments, the effector target is dickkopf-1 (DKK1 ) and the effector target binding domain comprises a DKK1 inhibitor (e.g., a DKK1 inhibitor provided in Jiang et al., Front Pharmacol, 13: Article No. 847387, 2022 or an antigen-binding fragment thereof).

In some embodiments, the effector target is receptor activator of nuclear factor kappa-p ligand (RANKL) and the effector target binding domain comprises a RANKL inhibitor (e.g., denosumab or an antigen-binding fragment thereof).

In some embodiments, the effector target is cathepsin K and the effector target binding domain comprises a cathepsin K inhibitor (e.g., odanacatib).

In some embodiments, the effector target is PTHR and the effector target binding domain comprises a PTHR agonist (e.g., an agonizing peptide, e.g., an intermittent agonizing peptide), e.g., to stimulate bone growth in osteoporosis.

Bone Address Target

An ANDbody of the invention also comprises an address target binder that binds to an address target to provide targeted delivery of the effector, wherein the address target is a bone tissue or cell. As used herein, an “address target” is a structure on a cell or tissue whose expression is sufficiently restricted in an organism to allow it to identify an organ, tissue, cell, or cell state of interest in an organism (e.g., a bone tissue or cell). The address target can be, e.g., a cell surface protein, or a structure localizing to the extracellular matrix. As used herein, “restricted” expression of an address target means that the address target has a differential, e.g., less broad, in vivo expression, as opposed to systemic expression. In certain embodiments, the address target is expressed, for example, in a single bone cell type, tissue or cell state in a mammalian subject, such as a human subject.

In some embodiments, the currently provided address target binding domains do not substantially influence biological signaling upon binding to the address target, e.g., does not modulate a signal transduction pathway or other biological response in the target bone cell or tissue. For example, the address target binder can be inert or inactive, in which it lacks any additional activity (other than binding), including lacking catalytic activity, after binding to the address target. For example, the address target binder binds a non-signaling site or motif of the address target. “Signal” is used herein to indicate a conformational, enzymatic, and/or electrical consequence occurs as a result of target binding. Accordingly, as described herein, address target binding domains do not signal upon address target binding. A domain that does not “substantially” influence biological signaling, as used herein, is a domain that modulates a signal transduction pathway or other biological response in the target cell or tissue to which it binds by no more than 25% relative to a control condition, e.g., relative to signaling in the absence of the domain. For example, the domain may modulate (e.g., increase or decrease) the signal transduction pathway or other biological response by less than 20%, less than 15%, less than 10%, less than 5%, less than 2%, or less than 1% (e.g., 20-25%, 15-20%, 10-15%, 5-10%, 2-5%, or 1 -2%). For example, in some embodiments, the address target binding domain does not substantially block chondrocyte alpha2 betal integrin dependent adhesion, e.g., decreases chondrocyte alpha2 betal integrin dependent adhesion by less than 20%, less than 15%, less than 10%, less than 5%, less than 2%, or less than 1% (e.g., 20-25%, 15-20%, 10-15%, 5-10%, 2-5%, or 1 -2%). In some embodiments, the address target binding domain does not substantially interfere with expression (e.g., upregulation) of one or more osteogenesis- related genes (e.g., Runx2, ALP and OCN), e.g., does not substantially interfere with upregulation of one or more osteogenesis-related genes in human osteosarcoma SaOS2 cells that have been subjected to osteogenic differentiation. In some embodiments, the address target binding domain does not substantially block mineralization by osteoblastic cells, e.g., blocks mineralization by less than 20%, less than 15%, less than 10%, less than 5%, less than 2%, or less than 1% (e.g., 20-25%, 15-20%, 10-15%, 5-10%, 2-5%, or 1 -2%).

Similarly, an effector target binding domain may not substantially signal, or may not signal at all, when it is not localized by an address target binding domain. In embodiments, an effector target binding domain signals with higher potency (e.g., has higher avidity) when it is localized to a bone tissue or cell by an address target binding domain compared to the signal when it is not localized by an address target binding domain. When an effector target binding domain is localized to a targeted bone cell or tissue by an address target binding domain as part of the same macromolecule, effector target signaling can be influenced as discussed above.

In some embodiments, the address target is used for organ-specific addressing (e.g., addressing to bones), tissue-specific addressing (e.g., addressing to one or more bone tissues), or cell-specific addressing (e.g., addressing to one or more bone cell types).

The specificity of address target binding domains for a cell or tissue can be detected using methods known in the art. In one embodiment, a Gini coefficient (GO) score, which is a method for assessing the expression variation of a particular gene in a data set, is used. (See O’Hagan et al., GeneGini: assessment via the Gini coefficient of reference “housekeeping” genes and diverse human transporter expression profiles. Cell systems 6, 230-244, https://doi.Org/10.1016/j. cels.2018.01 .003 (2018); Wright Muelas et al., The role and robustness of the Gini coefficient as an unbiased tool for the selection of Gini genes for normalising expression profiling data. Sci Rep 9, 17960 (2019). https://doi.org/10.1038/s41598-019-54288-7). Address target binders can be identified using cell expression data generated for address target binders as described herein (Table 2A and 2B). In some embodiments, address target markers exhibit Gini scores of greater than 0.4, such as between 0.74 and 1 .00. Conversely, non-address markers that are expressed more systemically exhibit Gini Scores of between 0.15 to 0.19.

In one embodiment, a Tau score, which represents the expression variation of a particular gene in a data set, is used. Calculating Tau uses the information of expression of a gene in each tissue and its maximal expression over all tissues while also taking into account the number of tissues where expression is measured (see Itai Yanai, et al., Genome-wide midrange transcription profiles reveal expression level relationships in human tissue specification, Bioinformatics, Volume 21 , Issue 5, 1 March 2005, Pages 650-659; Kryuchkova-Mostacci N, Robinson-Rechavi M. A benchmark of gene expression tissue-specificity metrics. Brief Bioinform. 2017 Mar 1 ;18(2):205-214. doi: 10.1093/bib/bbw008). In some embodiments, address target markers exhibit Tau scores of greater than 0.6, such as between 0.74 and 1 .00. Conversely, non-address markers that are expressed more systemically exhibit Tau Scores of below 0.3, such as 0.15 to 0.19.

In some embodiments, specificity of address target binding domains for a particular cell or tissue, such as that indicated by an appropriate Gini and/or Tau score, is determined with a tissue based analysis that does not include tissues having a natural biological separation barrier (i.e., bloodbrain barrier). For example, in some embodiments, Gini and/or Tau scores may be calculated without data from tissues such as (but not limited to): central nervous system, brain, eye, and/or testis tissues. In some embodiments, an address target as provided herein identifies a cell state. As used herein a “cell state” refers to a given physiological condition of a cell. A cell state may be, e.g., a disease state (relative to a non-disease state or normal state of a cell or tissue); or an activated state (relative to a non-activated state of a cell). Exemplary disease states include inflammation, infection (e.g., bacterial, viral, or fungal infection), and states relating to cancer (e.g., precancerous or cancerous cell states). In some aspects, cell state reflects the fact that cells of a particular type can exhibit variability with regard to one or more features and/or can exist in a variety of different conditions, while retaining the features of their particular cell type and not gain features that would cause them to be classified as a different cell type. The different states or conditions in which a cell can exist may be characteristic of a particular cell type (e.g., may involve properties or characteristics exhibited only by that cell type and/or involve functions performed only or primarily by that cell type) or may occur in multiple different cell types. In some embodiments, a cell state reflects the capability of a cell to respond to a particular stimulus or environmental condition (e.g., whether or not the cell will respond, or the type of response that will be elicited) or is a condition of the cell brought about by a stimulus or environmental condition. Cells in different cell states may be distinguished from one another in a variety of ways. For example, they may express, produce, or secrete one or more different genes, proteins, or other molecules ("markers", such as the address targets provided herein), exhibit differences in protein modifications such as phosphorylation, acetylation, etc., or may exhibit differences in appearance. Thus a cell state may be a condition of the cell in which the cell expresses, produces, or secretes one or more markers, exhibits particular protein modification(s), has a particular appearance, and/or will or will not exhibit one or more biological response(s) to a stimulus or environmental condition.

In some embodiments, the address target is dentin matrix acidic phosphoprotein 1 (DMP1 ).

In some embodiments, the address target is integrin binding sialoprotein (IBSP).

In some embodiments, the address target is trophoblast glycoprotein (TPBG).

In some embodiments, the address target is hydroxyapatite. In some embodiments, the address target is expressed on osteoblast cells, osteoclast cells, osteocytes, and/or bone lining cells. In some embodiments, the address target is expressed on osteoblast cells. In some embodiments, the address target is expressed on osteoclast cells. Exemplary address targets of the present technology are provided in Table 2, below.

Table 2: Exemplary Bone Address Targets

Small Molecules

A macromolecule of the invention (for example, an ANDbody) may be linked to a small molecule. The macromolecule and the small molecule may be linked by a cleavable linker. Alternatively, the macromolecule and the small molecule may be linked by a non-cleavable linker. Any useful linker may be employed for this purpose.

One or more (for example, one, two, three, four, five, or more) small molecules may be linked to the macromolecule. If multiple small molecules are linked to a macromolecule, the small molecules may be the same. Alternatively, one or more of the small molecules linked to the macromolecule may be different.

A small molecule to be linked to the macromolecule may be any desired small molecule. For example, the small molecule may be a therapeutic agent of interest that is to be localized or concentrated at a particular site by the macromolecule. In one example, the small molecule may be a therapeutic agent that acts together with or complements the effector target binding site domain. Alternatively, the small molecule may modulate the effector target binding site domain. In another example, the small molecule may modulate the address target binding site domain.

In some embodiments, the small molecule is a cathepsin K inhibitor, e.g., odanacatib.

In some embodiments, the small molecule is a bisphosphonate. In some embodiments, the bisphosphonate is alendronate, risedronate, etidronate, ibandronate, clodronate, tiludronate, pamidronate, or zoledronate. In some embodiments, the small molecule is alendronate.

Further exemplary classes of small molecules which may be linked to a macromolecule according to the invention include those in Table 3.

Table 3: Exemplary Small Molecule Classes

Particular examples of small molecules that may be linked to a macromolecule of the invention fall, for example, into any of the classes shown in Table 3. In particular, exemplary glucocorticoid receptor agonists include, without limitation, cortisone, dexamethasone, fluticasone, mometasone, fluocinolone, budesonide, butixicort, and betamethasone. Exemplary tyrosine protein kinase BTK inhibitors include, without limitation, acalabrutinib, evobrutinib, fenebrutinib, ibrutinib, orelabrutinib, pirtobrutinib, remibrutinib, rilzabrutinib, tolebrutinib, and zanubrutinib. Exemplary PI3K inhibitors include, without limitation, alpelisib, idelalisib, copanlisib, and duvelisib. Exemplary JAK inhibitors include, without limitation, abrocitinib, baricitinib, delgocitinib, filgotinib, peficitinib, ruxolitinib, tofacitinib, and upadacitinib. Exemplary cathepsin K inhibitors include, without limitation, odanacatib, relacatib, MIV-711 , and KGP-207. Exemplary topoisomerase inhibitors include, without limitation, irinotecan, doxorubicin, daunorubicin, doxorubicin, Ellence, etoposide, idarubicin, topotecan, and valrubicin.

Small molecules may be conjugated to a macromolecule of the invention using any conjugation technique known in the art. For example, small molecule carboxy, hydroxyl, and amine residues may be joined to amine and sulfhydryl residues on proteins using linkage techniques. Alternatively, any complementary functional groups on the two components may be used to react with each other to form a covalent bond. Examples of complementary reactive functional groups include, but are not limited to, e.g., maleimide and cysteine, amine and activated carboxylic acid, thiol and maleimide, activated sulfonic acid and amine, isocyanate and amine, azide and alkyne, and alkene and tetrazine. Further, any available linker may be utilized in the invention including heterobifunctional linkers that allow attachment of small molecules through, for example, disulfide bonds and amide bonds.

ANDbody Structures

In general, an ANDbody can be any macromolecule, such as a polypeptide or protein that contains both an effector target binding site or binding domain, and an address target binding site or binding domain. The binding sites may be present on the same polypeptide chain or different polypeptide chains that are linked together, e.g., through disulfide bonds.

In some embodiments, the binding site for the effector target and the binding site for the address target of the ANDbody each comprise an antibody heavy chain and/or a light chain domain. In some embodiments the ANDbody comprises a first antibody variable domain which has binding specificity for the effector target and a second antibody variable domain that has binding specificity for the address target. In other embodiments the ANDbody comprises a first antigen binding site of an antibody, which first antigen binding site has binding specificity for the effector target, and a second antigen binding site of an antibody, which second antigen binding site has binding specificity for the address target.

In some embodiments, the ANDbody may have the structure of an antibody molecule. The term “antibody” as used herein includes full-length antibodies and antigen binding antibody fragments (e.g., scFvs). In some embodiments, an antibody molecule has specificity for more than one, e.g., 2, 3, 4 antigens, e.g., the antibody molecule comprises a plurality of variable domain sequences, wherein a first variable domain sequence of the plurality has binding specificity for a first epitope {e.g., the effector target) and a second variable domain sequence of the plurality has binding specificity for a second epitope {e.g., the address target)

In some embodiments, the ANDbody is an antibody molecule that has an arm or domain that binds the effector target and an arm or domain that binds the address target. In embodiments, the ANDbody is an antibody molecule that comprises light chains that bind one of the effector target and address target, and heavy chains that bind the other of the effector target and address target.

In some embodiments, the ANDbody has the structure of an scFv, BsIgG, a BsAb fragment, a BiTE, a dual-affinity re-targeting protein (DART), a tandem diabody (TandAb), a diabody, an Fab2, a di-scFv, chemically linked F(ab’)2, an Ig molecule with 2, 3 or 4 different antigen binding sites, a DVI- IgG four-in-one, an ImmTac, an HSAbody, an IgG-IgG, a Cov-X-Body, an scFv1 -PEG-scFv2, an appended IgG, an DVD-IgG, an affibody, an affilin, an affimer, an affitin, an alphabody, an anticalin, an avimer, a DARPin, a Fynomer, a monobody, a nanoCLAMP, a bis-Fab, an Fv, a Fab, a Fab’-SH, a linear antibody, an scFv, an antibody with only a heavy chain (Humabody), an ScFab, an IgG antibody fragment, a single-chain variable region antibody, a single-domain heavy chain antibody, a bispecific triplebody, a BiKE, a CrossMAb, a dsDb, an scDb, tandem a dAb / VHH, a triple dAb VHH, a tetravalent dAb I VHH, a Fab-scFv, a Fab-Fv, or a DART-Fc, an adnectin, a Kunitz-type inhibitor, or a receptor decoy.

In some embodiments, the effector target binding site and/or the address target binding site comprises or consists of a small molecule. For example, in some embodiments, the first binding site comprises an antibody or antigen-binding fragment thereof and the second binding site comprises a small molecule that binds to the address target. In other embodiments, the first binding site comprises a small molecule that binds to the effector target and the second binding site comprises an antibody or antigen-binding fragment thereof. In aspects in which the ANDbody comprises a small molecule and a polypeptide component (e.g., an antibody or a fragment thereof), the small molecule may be conjugated to the polypeptide (e.g., antibody or fragment thereof) using any conjugation technique known in the art.

The affinity of the effector target binding site and address target binding site of an ANDbody for their respective binding partners may differ. In some embodiments the affinity of the first binding site to the therapeutic effector target it binds is weaker than the affinity of the second binding site to the address target. In some embodiments the affinity of the first binding site to the therapeutic effector target it binds is more than 2-fold, 3-fold, 4-fold, 5-fold, 10-fold, 15-fold, 20-fold, 50-fold, 100-fold, 200- fold, 500-fold, 1000-fold weaker than the affinity of the second binding site to the address target.

The terms "binding affinity" and “binding activity" refer to the tendency of a macromolecule, e.g., a polypeptide molecule, to bind or not to bind to a target. For purposes of the present invention, which combines two binding sites, the relative affinities of the two binding sites can be determined by, for example, measuring their respective affinities when each binding site is present on a common scaffold, such as in the form of a single chain antibody. Such a comparison allows a comparison of the affinities of two binding sites while eliminating any interference from other binding sites present on the macromolecule of the present invention.

Binding affinity may be quantified by determining the dissociation constant (Kd) for a polypeptide and its binder. A lower Kd is indicative of a higher affinity for a binding partner. Similarly, the specificity of binding of a polypeptide to its binding partner may be defined in terms of the comparative dissociation constants (Kd) of the polypeptide for its binding partner as compared to the dissociation constant with respect to the polypeptide and another, non-target molecule.

The value of this dissociation constant can be determined by known methods. For example, the Kd may be established using a double-filter nitrocellulose filter binding assay such as that disclosed by Wong & Lohman (Proc. Natl. Acad. Sci. USA 90, 5428-5432, 1993). Other standard assays to evaluate the binding ability of ligands such as antibodies towards targets are known in the art, including for example, ELISAs, Western blots, RIAs, and flow cytometry analysis. The binding kinetics (e.g., binding affinity) of the antibody also can be assessed by standard assays known in the art, such as by BiacoreTM system analysis. As an alternative to Kd, EC50 or IC50 may be used to determine relative affinities. In this context EC50 indicates the concentration at which a polypeptide achieves 50% of its maximum binding to a fixed quantity of binding partner. IC50 indicates the concentration at which a polypeptide inhibits 50% of the maximum binding of a fixed quantity of competitor to a fixed quantity of binding partner. In both cases, a lower level of EC50 or IC50 indicates a higher affinity for a target. The EC50 and IC50 values of an ANDbody binding site for its binding partner can both be determined by well-known methods, for example ELISA.

In some embodiments the Kd of therapeutic effector target binder might be higher than about 1 pM, about 10pM, about 10OpM, about 1 nM, about 10nM, about 10OnM, about 500nM, or about 1 uM (e.g., may be between 1 pM and 10pM, between 10 pM and 100pM, between 100 pM and 1 nM, between 1 nM and 10 nM, between 10 nM and 100 nM, between 100 nM and 500nM, or between 500 nM and 1 uM). In some embodiments the Kd of the address target binder might be less than about 1 uM, about 500nM, about 10OnM, about 10nM, about 1 nM, about 10OpM, about 10pM, or about 1 pM (e.g., may be between 1 uM and 500nM, between 500 nM and 10OnM, between 100 nM and 10nM, between 10 nM and 1 nM, between 1 nM and 10OpM, between 100 pM and 10 pM, or between 10pM and 1 pM). In some embodiments, the Kd for the therapeutic effector target binder may be about 6- fold, about 5-fold, about 4-fold, about 3-fold, or about 2-fold higher than the Kd for the address target binder.

In some embodiments the EC50 of therapeutic effector target binder might be higher than about 1 pM, about 10pM, about 10OpM, about 1 nM, about 10nM, about 10OnM, about 500nM, or about 1 uM (e.g., may be between 1 pM and 10pM, between 10 pM and 10OpM, between 100 pM and 1 nM, between 1 nM and 10 nM, between 10 nM and 100 nM, between 100 nM and 500nM, or between 500 nM and 1 uM). In some embodiments the EC50 of the address target binder might be less than about 1 uM, about 500nM, about 10OnM, about 10nM, about 1 nM, about 10OpM, about 10pM, or about 1 pM (e.g., may be between 1 uM and 500nM, between 500 nM and 10OnM, between 100 nM and 10nM, between 10 nM and 1 nM, between 1 nM and 10OpM, between 100 pM and 10 pM, or between 10pM and 1 pM). In some embodiments, the EC50 for the therapeutic effector target binder may be about 6- fold, about 5-fold, about 4-fold, about 3-fold, or about 2-fold higher than the EC50 for the address target binder.

In some embodiments the IC50 of therapeutic effector target binder might be higher than about 1 pM, about 10pM, about 10OpM, about 1 nM, about 10nM, about 10OnM, about 500nM, or about 1 uM (e.g., may be between 1 pM and 10pM, between 10 pM and 100pM, between 100 pM and 1 nM, between 1 nM and 10 nM, between 10 nM and 100 nM, between 100 nM and 500nM, or between 500 nM and 1 uM). In some embodiments the IC50 of the address target binder might be less than about 1 uM, about 500nM, about 10OnM, about 10nM, about 1 nM, about 10OpM, about 10pM, or about 1 pM (e.g., may be between 1 uM and 500nM, between 500 nM and 10OnM, between 100 nM and 10nM, between 10 nM and 1 nM, between 1 nM and 10OpM, between 100 pM and 10 pM, or between 10pM and 1 pM). In some embodiments, the IC50 for the therapeutic effector target binder may be about 6- fold, about 5-fold, about 4-fold, about 3-fold, or about 2-fold higher than the IC50 for the address target binder.

The cellular or tissue density of the effector target and address target bound by an ANDbody may differ. In embodiments, the density of the therapeutic effector target on a cell bound by the effector target binding site of an ANDbody is more than about 2-fold, about 3-fold, about 4-fold, about 5-fold, about 10-fold, about 15-fold, about 20-fold, about 50-fold, about 100-fold, about 200-fold, about 500-fold, about 1000-fold, about 10,000-fold, about 100,000-fold less than the density of the address target on a cell bound by the address target binding site.

In some embodiments, the affinity of the first binding site to the therapeutic effector target it binds is about one-half (¹/z) X Kd less than the affinity of the second binding site to the address target it binds and the density of the therapeutic effector target on a cell bound by the first binding site is about one-half (¹/z) X Kd less than the density of the address target on a cell bound by the second binding site.

In some embodiments, the ANDbody has both the affinity and density parameters as described hereinabove.

In some embodiments the first binding site and second binding site in the ANDbody are directly joined to each other. By directly joined is meant that the first binding site coding sequences abut the second binding site coding sequences and no sequences derived from other sequences (such as linkers) are present. In some embodiments the first binding site and second binding site in the ANDbody are not directly joined to each other.

An ANDbody as disclosed herein can also be linked to an additional moiety or moieties, e.g., an extracellular component, an intracellular component, a soluble factor {e.g., an enzyme, hormone, cytokine, growth factor, toxin, venom, pollutant, etc.), or a transmembrane protein {e.g., a cell surface receptor). In some embodiments, an ANDbody is linked to a small molecule and an additional moiety or moieties.

In some instances, effector target and/or address target sequences comprise full-length protein sequences and/or Fc fusion sequences with or without the signal peptide regions. In some embodiments, ANDbodies of the present technology include binding domains that bind bone address target or effector target proteins. In embodiments, binding domains of the present ANDbodies may bind protein sequences that include a signal peptide. In other embodiments, binding domains of the present ANDbodies may bind proteins that lack a signal protein. In some embodiments, binding domains of the present ANDbodies may bind full-length proteins. In other embodiments, binding domains of the present ANDbodies may bind protein fusions, such as full-length protein sequences, or peptide fragments thereof, with or without signal peptide regions, fused to other proteins, such as, for example, Fc sequences. In other embodiments, binding domains of the present ANDbodies may bind proteins that comprise less than the full-length protein sequence, such as a peptide fragment of the address target or effector target.

In some embodiments, an ANDbody of the invention comprises (a) a first binding site (effector target binding site) specific for sclerostin (SOST) and (b) a second binding site (address target binding site) specific for chondroadherin (CHAD). In some embodiments, an ANDbody of the invention comprises (a) a first binding site (effector target binding site) specific for sclerostin (SOST) and (b) a second binding site (address target binding site) specific for interferon-induced transmembrane protein 5 (IFITM5).

In some embodiments, an ANDbody of the invention comprises (a) a first binding site (effector target binding site) specific for Dickkopf-1 (DKK1 ) and (b) a second binding site (address target binding site) specific for chondroadherin (CHAD).

In some embodiments, an ANDbody of the invention comprises (a) a first binding site (effector target binding site) specific for sclerostin (SOST) and (b) a second binding site (address target binding site) specific for dentin matrix acidic phosphoprotein 1 (DMP1 ).

In some embodiments, an ANDbody of the invention comprises (a) a first binding site (effector target binding site) specific for sclerostin (SOST) and (b) a second binding site (address target binding site) specific for integrin binding sialoprotein (IBSP).

In some embodiments, an ANDbody of the invention comprises (a) a first binding site (effector target binding site) specific for sclerostin (SOST) and (b) a second binding site (address target binding site) specific for trophoblast glycoprotein (TPBG).

In some embodiments, an ANDbody of the invention comprises (a) a first binding site (effector target binding site) specific for sclerostin (SOST) and (b) a second binding site (address target binding site) specific for hydroxyapatite.

Production of ANDbody Compositions

Production of ANDbody polypeptides

ANDbody polypeptides of the invention may be produced by any suitable means. For example, all or part of the ANDbody may be expressed by a host cell comprising a nucleotide which encodes the ANDbody. Such methods of making a therapeutic polypeptide are routine in the art. See, in general, Smales & James (Eds.), Therapeutic Proteins: Methods and Protocols (Methods in Molecular Biology), Humana Press (2005); and Crommelin, Sindelar & Meibohm (Eds.), Pharmaceutical Biotechnology: Fundamentals and Applications, Springer (2013).

Methods for producing an ANDbody may involve expression in mammalian cells, although recombinant proteins can also be produced using insect cells, yeast, bacteria, or other cells under the control of appropriate promoters. Mammalian expression vectors may comprise nontranscribed elements such as an origin of replication, a suitable promoter and enhancer, and other 5' or 3' flanking nontranscribed sequences, and 5' or 3' nontranslated sequences such as necessary ribosome binding sites, a polyadenylation site, splice donor and acceptor sites, and termination sequences. DNA sequences derived from the SV40 viral genome, for example, SV40 origin, early promoter, enhancer, splice, and polyadenylation sites may be used to provide the other genetic elements required for expression of a heterologous DNA sequence. Appropriate cloning and expression vectors for use with bacterial, fungal, yeast, and mammalian cellular hosts are described in Green & Sambrook, Molecular Cloning: A Laboratory Manual (Fourth Edition), Cold Spring Harbor Laboratory Press (2012). Various mammalian cell culture systems can be employed to express and manufacture an ANDbody described herein. Examples of mammalian expression systems include CHO cells, COS cells, HeLA and BHK cell lines. Processes of host cell culture for production of protein therapeutics are described in, e.g., Zhou and Kantardjieff (Eds.), Mammalian Cell Cultures for Biologies Manufacturing (Advances in Biochemical Engineering/Biotechnology), Springer (2014). Purification of protein therapeutics is described in Franks, Protein Biotechnology: Isolation, Characterization, and Stabilization, Humana Press (2013); and in Cutler, Protein Purification Protocols (Methods in Molecular Biology), Humana Press (2010). Formulation of protein therapeutics is described in Meyer (Ed.), Therapeutic Protein Drug Products: Practical Approaches to formulation in the Laboratory, Manufacturing, and the Clinic, Woodhead Publishing Series (2012).

Antibody production techniques are known. See, for example, Zhiqiang (Editor), Therapeutic Monoclonal Antibodies: From Bench to Clinic. 1st Edition. Wiley 2009; Greenfield (Ed.) Antibodies: A Laboratory Manual. (Second edition) Cold Spring Harbor Laboratory Press 2013; Ferrara et al. 2012. Using Phage and Yeast Display to Select Hundreds of Monoclonal Antibodies: Application to Antigen 85, a Tuberculosis Biomarker. PLoS ONE 7(11 ): e49535, for methods of making recombinant antibodies, including antibody engineering, use of degenerate oligonucleotides, 5'-RACE, phage display, and mutagenesis; antibody testing and characterization; antibody pharmacokinetics and pharmacodynamics; antibody purification and storage; and screening and labeling techniques.

Production of ANDbody RNAs

In some embodiments, ANDbodies RNAs may be produced, e.g., for delivery to a subject. Generally, therapeutic mRNAs are made by in vitro transcription. Modification such as incorporation of modified bases, 5’cap analogues, and polyA tails can optimize activity and function. For example, translation and stability of mRNA can be accomplished, by cap and poly A tail modifications. E.g., incorporation of cap analogs such as ARCA (anti-reverse cap analogs) and a poly(A) tail of 100-200 bp into in vitro transcribed (IVT) mRNAs improves expression and stability (Kaczmarek et al. Genome Medicine (2017) 9:60). New types of cap analogs, such as 1 ,2-dithiodiphosphate-modified caps, can further improve efficiency of translation (Strenkowska et al. Nucleic Acids Res. 2016;44:9578-90). Codon optimization can also improve efficacy of protein synthesis and limit mRNA destabilization by rare codons (Presnyak et al. Cell. 2015;160:1111-24. 93; Thess et al. Mol Ther. 2015;23: 1456-64). Modifying 3' and 5' untranslated regions (UTRs), which contain sequences responsible for recruiting RNA-binding proteins (RBPs) and miRNAs, can enhance the level of protein product (Kaczmarek). Further, UTRs can be modified to encode regulatory elements (e.g., K-turn motifs and miRNA binding sites), in order to control RNA expression in a cell-specific manner (Wroblewska et al. Nat Biotechnol. 2015;33:839-41 ). RNA base modifications (e.g., pseudouridine incorporated mRNA, e.g., N1 -methyl- pseudouridine) contribute to masking mRNA immune-stimulatory activity and increase mRNA translation by enhancing translation initiation (Andries et al. J Control Release. 2015;217:337-44; Svitkin et al. Nucleic Acids Res. 2017;45:6023-36). mRNA compositions and methods of their manufacture are known and are disclosed, e.g., in WO2016011306; WO2016014846; WO2016022914; WO2016077123; WO2016164762; WO2016201377; WO2017049275; US9937233; US8710200; US10022425; US9878056; US9572897; Jemiel ity et al. RNA. 2003;9:1108-22. 90; Mockey et al. Biochem Biophys Res Commun. 2006;340:1062-8. 91 ; Strenkowska et al. Nucleic Acids Res. 2016;44:9578-90. 92; Presnyak et al. Cell. 2015;160:1111-24. 93; Kaczmarek et al. Genome Medicine (2017) 9:60.

Production Of ANDbodies With Altered Affinities

ANDbodies with binding sites with altered affinities can be made using methods known in the art, e.g., an ANDbody can be engineered to have a target binding site that has decreased affinity for the effector target. See, e.g., US Patent No. 10,654,928. In general, an ANDBody may be modified to alter the affinity of an effector target binding site to its effector target or to alter the affinity of an address target binding site to its address target. The modification can increase or decrease affinity for the binding site’s binding partner.

Assessment of Targets and Addresses

Expression of a therapeutic target can be assessed at either the RNA or protein level using methods known in the art. In embodiments, expression of the therapeutic target is assessed by measuring RNA expression, e.g., using an RNA sequence dataset as a proxy for protein expression levels. RNA datasets include those a genotype-Tissue Expression (GTEx) dataset (see, e.g., https://www.genome.gov/Funded-Programs-Projects/Genotype-Tissue-Expression-Project) or a Human Protein Atlas (HPA) dataset (https://www.proteinatlas.org/).

A non-limiting list of tissues in which expression of the therapeutic target can be assessed includes, e.g., bone tissues and/or bone cells, the minor salivary gland, thyroid, lung, breast (mammary tissue), pancreas, adrenal gland, liver, kidney (cortex), kidney (medulla), adipose-viscaral (omentum), small intestine - terminal ileum, fallopian tube, ovary, uterus, skin not sun exposed (suprapubic); cervix — endocervix, cervix — ectocervix, vagina, skin sun exposed (lower leg), cells eneanterior cingulate cortex (BA24), caudate (basal ganglia), putamen (basal ganglia), nucleus acumbens (basal ganglia), hypothalamus, amygdala, hippocampus, cerebellum/cerebellar hemisphere, substantia nigra, pituitary, spinal cord (cervical), artery-aorta, heart-atrial appendage, artery-coronary- heart, left ventricle, esophagus-mucosa, esophagus-muscularis, esophagus-gastroesophageal junction, spleen, stomach, colon-transverse, colon — sigmoid, testis, whole blood, cells - (EBV-transformed lymphocytes, artery-tibial, or nerve-tibial tissues. In some embodiments, expression of the address target is substantially higher in bone tissue than in any other tissue. In some embodiments, expression of the address target is restricted to bone tissue.

Address markers can be assessed using methods well known in the art, e.g., gene expression can be assessed at the mRNA level using Northern blots, cDNA or oligonucleotide microarrays, or sequencing {e.g., RNA-Seq), or at the level of protein expression using protein microarrays, Western blots, flow cytometry, immunohistochemistry, etc. Modifications can be assessed, e.g., using antibodies that are specific for a particular modified form of a protein, e.g., phospho-specific antibodies, or mass spectrometry.

Uses of ANDbodies

ANDbodies and their pharmaceutical compositions provided herein are suitable for administration to a subject in need thereof, wherein the subject is a human or a non-human animal, for example, suitable for human therapeutic or veterinary use.

Veterinary use includes use for treatment of mammals, including commercially relevant mammals, e.g., pet and live-stock animals, such as cattle, pigs, horses, sheep, goats, cats, dogs, mice, and/or rats; and/or birds, including commercially relevant birds such as parrots, poultry, chickens, ducks, geese, hens or roosters and/or turkeys; zoo animals, e.g., a feline; non-mammal animals, e.g., reptiles, fish, amphibians, etc.

The invention is further directed to a subject or subject cell comprising the ANDbody composition described herein. In some embodiments, the subject or subject cell is a plant, insect, bacteria, fungus, vertebrate, mammal {e.g., human), or other organism or cell.

In some embodiments, a subject or a subject cell is contacted with {e.g., delivered to or administered to) the ANDbody composition. In some embodiments, the subject is a mammal, such as a human. The amount of the ANDbody composition, expression product, or both in the subject can be measured at any time after administration.

In some embodiments, a subject to which an ANDbody or ANDbody composition provided herein is administered has a disease, disorder, or condition of the bone. For example, in some embodiments, the subject has a bone density disease (e.g., low bone density, osteopenia, or osteoporosis), a kidney disease tied to calcium leaching from bones (e.g., renal bone disease, chronic kidney disease (CKD) (e.g., CKD associated with mineral and bone disorder), or renal osteodystrophy), osteoarthritis, rheumatoid arthritis, osteogenesis imperfecta, Paget’s disease, fibrous dysplasia, osteomalacia, rickets, osteonecrosis, a bone cancer (e.g., osteosarcoma or Ewing sarcoma), osteomyelitis, or a bone infection.

Pharmaceutical Compositions

Polypeptide Pharmaceutical Compositions

The ANDbody compositions described herein {e.g., ANDbody polypeptide or RNA compositions) may be administered to a subject in need thereof. The invention includes pharmaceutical compositions that include an ANDbody composition in combination with one or more pharmaceutically acceptable excipients.

Formulation of protein therapeutics is routine. See, for example, Ribeiro et al., Insights on the Formulation of Recombinant Proteins. Adv Biochem Eng Biotechnol. 2020;171 :23-54. doi: 10.1007/10_2019_119. PMID: 31844925.

RNA pharmaceutical compositions

Nucleic acids {e.g., RNA) encoding an ANDBody can alternatively or additionally be administered to a subject. Generally, therapeutic mRNAs are made by in vitro transcription. Modification such as incorporation of modified bases, 5’cap analogues, and polyA tails can optimize activity and function. For example, translation and stability of mRNA can be accomplished, by cap and poly A tail modifications. E.g., incorporation of cap analogs such as ARCA (anti-reverse cap analogs) and a poly(A) tail of 100-200 bp into in vitro transcribed (IVT) mRNAs improves expression and stability (Kaczmarek et al. Genome Medicine (2017) 9:60). New types of cap analogs, such as 1 ,2- dithiodiphosphate-modified caps, can further improve efficiency of translation (Strenkowska et al. Nucleic Acids Res. 2016;44:9578-90). Codon optimization can also improve efficacy of protein synthesis and limit mRNA destabilization by rare codons (Presnyak et al. Cell. 2015;160:1111-24. 93; Thess et al. Mol Ther. 2015;23: 1456-64). Modifying 3' and 5' untranslated regions (UTRs), which contain sequences responsible for recruiting RNA-binding proteins (RBPs) and miRNAs, can enhance the level of protein product (Kaczmarek). Further, UTRs can be modified to encode regulatory elements (e.g., K-turn motifs and miRNA binding sites), in order to control RNA expression in a cell-specific manner (Wroblewska et al. Nat Biotechnol. 2015;33:839-41 ). RNA base modifications (e.g., pseudouridine incorporated mRNA, e.g., N1 -methyl-pseudouridine) contribute to masking mRNA immune-stimulatory activity and increase mRNA translation by enhancing translation initiation (Andries et al. J Control Release. 2015;217:337-44; Svitkin et al. Nucleic Acids Res. 2017;45:6023-36). mRNA compositions and methods of their manufacture are known and are disclosed, e.g., in WO2016011306; WO2016014846; WO2016022914; WO2016077123; WO2016164762; WO2016201377;

WO2017049275; US9937233; US8710200; US10022425; US9878056; US9572897; Jemielity et al. RNA. 2003;9:1108-22. 90; Mockey et al. Biochem Biophys Res Commun. 2006;340:1062-8. 91 ; Strenkowska et al. Nucleic Acids Res. 2016;44:9578-90. 92; Presnyak et al. Cell. 2015;160:1111 -24. 93; Kaczmarek et al. Genome Medicine (2017) 9:60.

In embodiments, the RNA is a circular RNA. See, for example, WO2019118919, describing the expression of a therapeutic RNA, such as an antibody RNA, from a circular RNA. In some embodiments, the invention includes a circular polyribonucleotide that comprises (a) an internal ribosome entry site (IRES), (b) an expression sequence encoding a ANDbody described herein and lacking a poly-A sequence, and (c) a termination element. A circular RNA encoding an ANDbody described herein may be delivered naked (i.e., without formulation with a carrier) or with a carrier.

Combination Therapies

In some embodiments, an ANDbody or ANDbody composition provided herein is administered in combination with one or more additional therapeutic agents, e.g., one or more additional bone therapeutic agents. For example, the ANDbody or ANDbody composition may be administered in combination with an agent used for the treatment of a bone density disease (e.g., low bone density, osteopenia, or osteoporosis), a kidney disease tied to calcium leaching from bones (e.g., renal bone disease, CKD (e.g., CKD associated with mineral and bone disorder), or renal osteodystrophy), osteoarthritis, rheumatoid arthritis, osteogenesis imperfecta, Paget’s disease, fibrous dysplasia, osteomalacia, rickets, osteonecrosis, a bone cancer (e.g., osteosarcoma or Ewing sarcoma), osteomyelitis, or a bone infection.

In some embodiments, the additional bone therapeutic agent is an agent used for the treatment of osteoporosis, e.g., a bisphosphonate (e.g., alendronate, ibandronate, risedronate, or zoledronic acid).

In some embodiments, the additional bone therapeutic agent is a sclerostin (SOST) inhibitor (e.g., romosozumab, blosozumab, setrusumab, SHR-1222, or a SOST inhibitor. See, e.g., Yu et al., Acta Pharm Sin B, 12(5): 2150-2170, 2022), a dickkopf-1 (DKK1 ) inhibitor (see, e.g., Jiang et al., Front Pharmacol, 13: Article No. 847387, 2022), a receptor activator of nuclear factor kappa-p ligand (RANKL) inhibitor (e.g., denosumab), or a cathepsin K inhibitor (e.g., odanacatib).

Carriers

Lipid nanoparticles

Formulations of the compositions described herein (e.g., polypeptide or RNA ANDbody compositions) for in vivo delivery with a carrier include lipid nanoparticle (LNP) formulations. See, e.g., US Pat. 9,764,036; US Pat. 9,682,139; Kauffman et al. Nano Lett. 2015;15: 7300-6. 37; Fenton et al. Adv Mater. 2016;28:2939-43). LNPs, in some embodiments, comprise one or more ionic lipids, such as non-cationic lipids (e.g., neutral or anionic, or zwitterionic lipids); one or more conjugated lipids (such as PEG-conjugated lipids or lipids conjugated to polymers described in Table 5 of WO2019217941 ; incorporated herein by reference in its entirety); one or more sterols (e.g., cholesterol); and, optionally, one or more targeting molecules (e.g., conjugated receptors, receptor ligands, antibodies); or combinations of the foregoing.

Lipids that can be used in nanoparticle formations {e.g., lipid nanoparticles) include, for example those described in Table 4 of WO2019217941 , which is incorporated herein by reference — e.g., a lipid-containing nanoparticle can comprise one or more of the lipids in Table 4 of WO2019217941 . Lipid nanoparticles can include additional elements, such as polymers, such as the polymers described in Table 5 of WO2019217941 , incorporated by reference.

In some embodiments, conjugated lipids, when present, can include one or more of PEG- diacylglycerol (DAG) (such as l-(monomethoxy-polyethyleneglycol)-2,3- dimyristoylglycerol (PEG- DMG)), PEG-dialkyloxypropyl (DAA), PEG-phospholipid, PEG- ceramide (Cer), a pegylated phosphatidylethanoloamine (PEG-PE), PEG succinate diacylglycerol (PEGS-DAG) (such as 4-0-(2',3'- di(tetradecanoyloxy)propyl-l-0-(w- methoxy(polyethoxy)ethyl) butanedioate (PEG-S-DMG)), PEG dialkoxypropylcarbam, N- (carbonyl-methoxypoly ethylene glycol 2000)- 1 ,2-distearoyl-sn-glycero-3- phosphoethanolamine sodium salt, and those described in Table 2 of WO2019051289 (incorporated by reference), and combinations of the foregoing.

In some embodiments, sterols that can be incorporated into lipid nanoparticles include one or more of cholesterol or cholesterol derivatives, such as those in W02009/127060 or US2010/0130588, which are incorporated by reference. Additional exemplary sterols include phytosterols, including those described in Eygeris et al (2020), dx.doi.org/10.1021/acs.nanolett.0c01386, incorporated herein by reference.

In some embodiments, the lipid particle comprises an ionizable lipid, a non-cationic lipid, a conjugated lipid that inhibits aggregation of particles, and a sterol. The amounts of these components can be varied independently and to achieve desired properties. For example, in some embodiments, the lipid nanoparticle comprises an ionizable lipid is in an amount from about 20 mol % to about 90 mol % of the total lipids (in other embodiments it may be 20-70% (mol), 30-60% (mol) or 40-50% (mol); about 50 mol % to about 90 mol % of the total lipid present in the lipid nanoparticle), a non-cationic lipid in an amount from about 5 mol % to about 30 mol % of the total lipids, a conjugated lipid in an amount from about 0.5 mol % to about 20 mol % of the total lipids, and a sterol in an amount from about 20 mol % to about 50 mol % of the total lipids. The ratio of total lipid to nucleic acid can be varied as desired. For example, the total lipid to nucleic acid (mass or weight) ratio can be from about 10: 1 to about 30: 1 .

In some embodiments, the lipid to nucleic acid ratio (mass/mass ratio; w/w ratio) can be in the range of from about 1 : 1 to about 25: 1 , from about 10: 1 to about 14: 1 , from about 3 : 1 to about 15: 1 , from about 4: 1 to about 10: 1 , from about 5: 1 to about 9: 1 , or about 6: 1 to about 9: 1 . The amounts of lipids and nucleic acid can be adjusted to provide a desired N/P ratio, for example, N/P ratio of 3, 4, 5, 6, 7, 8, 9, 10 or higher. Generally, the lipid nanoparticle formulation’s overall lipid content can range from about 5 mg/ml to about 30 mg/mL.

Some non-limiting example of lipid compounds that may be used (e.g., in combination with other lipid components) to form lipid nanoparticles for the delivery of compositions described herein, e.g., nucleic acid (e.g., RNA) described herein includes,

In some embodiments an LNP comprising Formula (i) is used to deliver an ANDbody RNA composition described herein to the liver and/or hepatocyte cells.

In some embodiments an LNP comprising Formula (ii) is used to deliver an ANDbody RNA composition described herein to the liver and/or hepatocyte cells.

In some embodiments an LNP comprising Formula (Hi) is used to deliver an ANDbody RNA composition described herein to the liver and/or hepatocyte cells.

In some embodiments an LNP comprising Formula (v) is used to deliver an ANDbody RNA composition described herein to the liver and/or hepatocyte cells.

In some embodiments an LNP comprising Formula (vi) is used to deliver an ANDbody RNA composition described herein to the liver and/or hepatocyte cells.

(vii)

(viii)

In some embodiments an LNP comprising Formula (viii) is used to deliver an ANDbody RNA composition described herein to the liver and/or hepatocyte cells.

In some embodiments an LNP comprising Formula (ix) is used to deliver an ANDbody RNA composition described herein to the liver and/or hepatocyte cells.

wherein

X1 is O, NR1 , or a direct bond, X2 is C2-5 alkylene, X3 is C(=O) or a direct bond, R1 is H or Me, R3 is Ci-3 alkyl, R2 is Ci-3 alkyl, or R2 taken together with the nitrogen atom to which it is attached and 1 -3 carbon atoms of X2 form a 4-, 5-, or 6-membered ring, or X1 is NR1 , R1 and R2 taken together with the nitrogen atoms to which they are attached form a 5- or 6-membered ring, or R2 taken together with R3 and the nitrogen atom to which they are attached form a 5-, 6-, or 7-membered ring, Y1 is C2- 12 alkylene, Y2 is selected from

, (in either orientation),

(in either orientation), n is 0 to 3, R4 is Ci-15 alkyl, Z1 is Ci-6 alkylene or a direct bond,

(in either orientation) or absent, provided that if Z1 is a direct bond,

Z2 is absent;

R5 is C5-9 alkyl or C6-10 alkoxy, R6 is C5-9 alkyl or C6-10 alkoxy, W is methylene or a direct bond, and R7 is H or Me, or a salt thereof, provided that if R3 and R2 are C2 alkyls, X1 is O, X2 is linear C3 alkylene, X3 is C(=0), Y1 is linear Ce alkylene, (Y2 )n-R4 is

R4 is linear C5 alkyl, Z1 is C2 alkylene, Z2 is absent, W is methylene, and R7 is H, then R5 and R6 are not Cx alkoxy.

In some embodiments an LNP comprising Formula (xii) is used to deliver an ANDbody RNA composition described herein to the liver and/or hepatocyte cells.

In some embodiments an LNP comprising Formula (xi) is used to deliver an ANDbody RNA composition described herein to the liver and/or hepatocyte cells.

(xiv)

In some embodiments an LNP comprises a compound of Formula (xiii) and a compound of Formula (xiv).

In some embodiments an LNP comprising Formula (xv) is used to deliver an ANDbody RNA composition described herein to the liver and/or hepatocyte cells.

(xvi)

In some embodiments an LNP comprising a formulation of Formula (xvi) is used to deliver an ANDbody RNA composition described herein to the lung endothelial cells.

(xviii) (a)

In some embodiments, a lipid compound used to form lipid nanoparticles for the delivery of compositions described herein, e.g., nucleic acid (e.g., RNA) described herein is made by one of the following reactions:

(XX) (a)

(xx)(b)

In some embodiments, a composition described herein (e.g., a nucleic acid or a protein) is provided in an LNP that comprises an ionizable lipid. In some embodiments, the ionizable lipid is heptadecan-9-yl 8-((2-hydroxyethyl)(6-oxo-6-(undecyloxy)hexyl)amino)octanoate (SM-102); e.g., as described in Example 1 of US9,867,888 (incorporated by reference herein in its entirety). In some embodiments, the ionizable lipid is 9Z,12Z)-3-((4,4-bis(octyloxy)butanoyl)oxy)-2-((((3- (diethylamino)propoxy)carbonyl)oxy)methyl)propyl octadeca-9,12-dienoate (LP01 ), e.g., as synthesized in Example 13 of WO2015/095340 (incorporated by reference herein in its entirety). In some embodiments, the ionizable lipid is Di((Z)-non-2-en-1 -yl) 9-((4- dimethylamino)butanoyl)oxy)heptadecanedioate (L319), e.g. as synthesized in Example 7, 8, or 9 of US2012/0027803 (incorporated by reference herein in its entirety). In some embodiments, the ionizable lipid is 1 ,1 '-((2-(4-(2-((2-(Bis(2-hydroxydodecyl)amino)ethyl)(2-hydroxydodecyl) amino)ethyl)piperazin-1 - yl)ethyl)azanediyl)bis(dodecan-2-ol) (C12-200), e.g., as synthesized in Examples 14 and 16 of WO2010/053572 (incorporated by reference herein in its entirety). In some embodiments, the ionizable lipid is Imidazole cholesterol ester (ICE) lipid (3S, 10R, 13R, 17R)-10, 13-dimethyl-17- ((R)-6- methylheptan-2-yl)-2, 3, 4, 7, 8, 9, 10, 11 , 12, 13, 14, 15, 16, 17-tetradecahydro-IH- cyclopenta[a]phenanthren-3-yl 3-(1 H-imidazol-4-yl)propanoate, e.g., Structure (I) from W02020/106946 (incorporated by reference herein in its entirety).

In some embodiments, an ionizable lipid may be a cationic lipid, an ionizable cationic lipid, e.g., a cationic lipid that can exist in a positively charged or neutral form depending on pH, or an amine- containing lipid that can be readily protonated. In some embodiments, the cationic lipid is a lipid capable of being positively charged, e.g., under physiological conditions. Exemplary cationic lipids include one or more amine group(s) which bear the positive charge. In some embodiments, the lipid particle comprises a cationic lipid in formulation with one or more of neutral lipids, ionizable amine-containing lipids, biodegradable alkyne lipids, steroids, phospholipids including polyunsaturated lipids, structural lipids {e.g., sterols), PEG, cholesterol and polymer conjugated lipids. In some embodiments, the cationic lipid may be an ionizable cationic lipid. An exemplary cationic lipid as disclosed herein may have an effective pKa over 6.0. In embodiments, a lipid nanoparticle may comprise a second cationic lipid having a different effective pKa (e.g., greater than the first effective pKa), than the first cationic lipid. A lipid nanoparticle may comprise between 40 and 60 mol percent of a cationic lipid, a neutral lipid, a steroid, a polymer conjugated lipid, and a therapeutic agent, e.g., a nucleic acid (e.g., RNA) described herein, encapsulated within or associated with the lipid nanoparticle. In some embodiments, the nucleic acid is co-formulated with the cationic lipid. The nucleic acid may be adsorbed to the surface of an LNP, e.g., an LNP comprising a cationic lipid. In some embodiments, the nucleic acid may be encapsulated in an LNP, e.g., an LNP comprising a cationic lipid. In some embodiments, the lipid nanoparticle may comprise a targeting moiety, e.g., coated with a targeting agent. In embodiments, the LNP formulation is biodegradable. In some embodiments, a lipid nanoparticle comprising one or more lipid described herein, e.g., Formula (i), (ii), (ii), (vii) and/or (ix) encapsulates at least 1%, at least 5%, at least 10%, at least 20%, at least 30%, at least 40%, at least 50%, at least 60%, at least 70%, at least 80%, at least 90%, at least 92%, at least 95%, at least 97%, at least 98% or 100% of an RNA molecule.

Exemplary ionizable lipids that can be used in lipid nanoparticle formulations include, without limitation, those listed in Table 1 of WO2019051289, incorporated herein by reference. Additional exemplary lipids include, without limitation, one or more of the following formulae: X of

US2013/0090372; A of US2013/0274523; A of US2013/0274504; A of US2013/0053572; A of W02013/016058; A of W02012/162210; I of US2008/042973; I, II, III, or IV of US2012/01287670; I or II of US2014/0200257; I, II, or III of US2015/0203446; I or III of US2015/0005363; I, IA, IB, IC, ID, II, HA, IIB, IIC, HD, or lll-XXIV of US2014/0308304; of US2013/0338210; I, II, III, or IV of W02009/132131 ; A of US2012/0101 1478; I or XXXV of US2012/0027796; XIV or XVII of US2012/0058144; of

US2013/0323269; I of US201 1/01 17125; I, II, or III of US201 1/0256175; I, II, III, IV, V, VI, VII, VIII, IX, X, XI, XII of US2012/0202871 ; I, II, III, IV, V, VI, VII, VIII, X, XII, XIII, XIV, XV, or XVI of US201 1/0076335; I or II of US2006/008378; I of US2013/0123338; I or X-A-Y-Z of US2015/0064242; XVI, XVII, or XVIII of US2013/0022649; I, II, or III of US2013/01 16307; I, II, or III of US2013/01 16307; I or II of US2010/0062967; l-X of US2013/0189351 ; I of US2014/0039032; V of US2018/0028664; I of

US2016/0317458; I of US2013/0195920; 5, 6, or 10 of US10,221 , 127; HI-3 of WO2018/081480; I-5 or I- 8 of W02020/081938; 18 or 25 of US9,867,888; A of US2019/0136231 ; II of WO2020/219876; 1 of US2012/0027803; OF-02 of US2019/0240349; 23 of US10,086,013; cKK-E12/A6 of Miao et al (2020); C12-200 of WO2010/053572; 7C1 of Dahlman et al (2017); 304-013 or 503-013 of Whitehead et al ; TS-P4C2 of US9,708,628; I of W02020/106946; I of W02020/106946.

In some embodiments, the ionizable lipid is MC3 (6Z,9Z,28Z,3 IZ)-heptatriaconta- 6,9,28,3 I- tetraen-l9-yl-4-(dimethylamino) butanoate (DLin-MC3-DMA or MC3), e.g., as described in Example 9 of WO2019051289A9 (incorporated by reference herein in its entirety). In some embodiments, the ionizable lipid is the lipid ATX-002, e.g., as described in Example 10 of WO2019051289A9 (incorporated by reference herein in its entirety). In some embodiments, the ionizable lipid is (I3Z,I6Z)- A,A-dimethyl-3- nonyldocosa-13, 16-dien-l-amine (Compound 32), e.g., as described in Example 1 1 of WO2019051289A9 (incorporated by reference herein in its entirety). In some embodiments, the ionizable lipid is Compound 6 or Compound 22, e.g., as described in Example 12 of WO2019051289A9 (incorporated by reference herein in its entirety).

Exemplary non-cationic lipids include, but are not limited to, distearoyl-sn-glycero- phosphoethanolamine, distearoylphosphatidylcholine (DSPC), dioleoylphosphatidylcholine (DOPC), dipalmitoylphosphatidylcholine (DPPC), dioleoylphosphatidylglycerol (DOPG), dipalmitoylphosphatidylglycerol (DPPG), dioleoyl-phosphatidylethanolamine (DOPE), palmitoyloleoylphosphatidylcholine (POPC), palmitoyloleoylphosphatidylethanolamine (POPE), dioleoylphosphatidylethanolamine 4-(N-maleimidomethyl)-cyclohexane- 1 - carboxylate (DOPE-mal), dipalmitoyl phosphatidyl ethanolamine (DPPE), dimyristoylphosphoethanolamine (DMPE), distearoyl- phosphatidyl-ethanolamine (DSPE), monomethyl-phosphatidylethanolamine (such as 16-0- monomethyl PE), dimethyl- phosphatidylethanolamine (such as 16-O-dimethyl PE), 18-l-trans PE, I- stearoyl-2-oleoyl- phosphatidyethanolamine (SOPE), hydrogenated soy phosphatidylcholine (HSPC), egg phosphatidylcholine (EPC), dioleoylphosphatidylserine (DOPS), sphingomyelin (SM), dimyristoyl phosphatidylcholine (DMPC), dimyristoyl phosphatidylglycerol (DMPG), distearoylphosphatidylglycerol (DSPG), dierucoylphosphatidylcholine (DEPC), palmitoyloleyolphosphatidylglycerol (POPG), dielaidoylphosphatidylethanolamine (DEPE), lecithin, phosphatidylethanolamine, lysolecithin, lysophosphatidylethanolamine, phosphatidylserine, phosphatidylinositol, sphingomyelin, egg sphingomyelin (ESM), cephalin, cardiolipin, phosphatidicacid, cerebrosides, dicetylphosphate, lysophosphatidylcholine, dilinoleoylphosphatidylcholine, or mixtures thereof. It is understood that other diacylphosphatidylcholine and diacylphosphatidylethanolamine phospholipids can also be used. The acyl groups in these lipids are preferably acyl groups derived from fatty acids having C10-C24 carbon chains, e.g., lauroyl, myristoyl, palmitoyl, stearoyl, or oleoyl. Additional exemplary lipids, in certain embodiments, include, without limitation, those described in Kim et al. (2020) dx.doi.org/10.1021/acs.nanolett.0c01386, incorporated herein by reference. Such lipids include, in some embodiments, plant lipids found to improve liver transfection with mRNA (e.g., DGTS).

Other examples of non-cationic lipids suitable for use in the lipid nanoparticles include, without limitation, nonphosphorous lipids such as, e.g., stearylamine, dodeeylamine, hexadecylamine, acetyl palmitate, glycerol ricinoleate, hexadecyl stereate, isopropyl myristate, amphoteric acrylic polymers, triethanolamine-lauryl sulfate, alkyl-aryl sulfate polyethyloxylated fatty acid amides, dioctadecyl dimethyl ammonium bromide, ceramide, sphingomyelin, and the like. Other non-cationic lipids are described in WO2017/099823 or US patent publication US2018/0028664, the contents of which is incorporated herein by reference in their entirety.

In some embodiments, the non-cationic lipid is oleic acid or a compound of Formula I, II, or IV of US2018/0028664, incorporated herein by reference in its entirety. The non-cationic lipid can comprise, for example, 0-30% (mol) of the total lipid present in the lipid nanoparticle. In some embodiments, the non-cationic lipid content is 5-20% (mol) or 10-15% (mol) of the total lipid present in the lipid nanoparticle. In embodiments, the molar ratio of ionizable lipid to the neutral lipid ranges from about 2:1 to about 8:1 (e.g., about 2:1 , 3:1 , 4:1 , 5:1 , 6:1 , 7:1 , or 8:1 ).

In some embodiments, the lipid nanoparticles do not comprise any phospholipids.

In some aspects, the lipid nanoparticle can further comprise a component, such as a sterol, to provide membrane integrity. One exemplary sterol that can be used in the lipid nanoparticle is cholesterol and derivatives thereof. Non-limiting examples of cholesterol derivatives include polar analogues such as 5a-choiestanol, 53-coprostanol, choiesteryl-(2,-hydroxy)-ethyl ether, choiesteryl-(4'- hydroxy)-butyl ether, and 6-ketocholestanol; non-polar analogues such as 5a-cholestane, cholestenone, 5a-cholestanone, 5p-cholestanone, and cholesteryl decanoate; and mixtures thereof. In some embodiments, the cholesterol derivative is a polar analogue, e.g., choiesteryl-(4 '-hydroxy)-buty1 ether. Exemplary cholesterol derivatives are described in PCT publication W02009/127060 and US patent publication US2010/0130588, each of which is incorporated herein by reference in its entirety.

In some embodiments, the component providing membrane integrity, such as a sterol, can comprise 0-50% (mol) (e.g., 0-10%, 10-20%, 20-30%, 30-40%, or 40-50%) of the total lipid present in the lipid nanoparticle. In some embodiments, such a component is 20-50% (mol) 30-40% (mol) of the total lipid content of the lipid nanoparticle.

In some embodiments, the lipid nanoparticle can comprise a polyethylene glycol (PEG) or a conjugated lipid molecule. Generally, these are used to inhibit aggregation of lipid nanoparticles and/or provide steric stabilization. Exemplary conjugated lipids include, but are not limited to, PEG-lipid conjugates, polyoxazoline (POZ)-lipid conjugates, polyamide-lipid conjugates (such as ATTA-lipid conjugates), cationic-polymer lipid (CPL) conjugates, and mixtures thereof. In some embodiments, the conjugated lipid molecule is a PEG-lipid conjugate, for example, a (methoxy polyethylene glycol)- conjugated lipid.

Exemplary PEG-lipid conjugates include, but are not limited to, PEG-diacylglycerol (DAG) (such as l-(monomethoxy-polyethyleneglycol)-2,3-dimyristoylglycerol (PEG-DMG)), PEG-dialkyloxypropyl (DAA), PEG-phospholipid, PEG-ceramide (Cer), a pegylated phosphatidylethanoloamine (PEG-PE), PEG succinate diacylglycerol (PEGS-DAG) (such as 4-0-(2',3'-di(tetradecanoyloxy)propyl-l-0-(w- methoxy(polyethoxy)ethyl) butanedioate (PEG-S-DMG)), PEG dialkoxypropylcarbam, N-(carbonyl- methoxypolyethylene glycol 2000)-l,2-distearoyl-sn-glycero-3-phosphoethanolamine sodium salt, or a mixture thereof. Additional exemplary PEG-lipid conjugates are described, for example, in US5,885,6I3, US6,287,59I,

US2003/0077829, US2003/0077829, US2005/0175682, US2008/0020058, US2011/0117125, US2010/0130588, US2016/0376224, US2017/0119904, and US/099823, the contents of all of which are incorporated herein by reference in their entirety. In some embodiments, a PEG-lipid is a compound of Formula III, lll-a-l, lll-a-2, lll-b-1 , lll-b-2, or V of US2018/0028664, the content of which is incorporated herein by reference in its entirety. In some embodiments, a PEG-lipid is of Formula II of US20150376115 or US2016/0376224, the content of both of which is incorporated herein by reference in its entirety. In some embodiments, the PEG-DAA conjugate can be, for example, PEG- dilauryloxypropyl, PEG- dimyristyloxypropyl, PEG-dipalmityloxypropyl, or PEG-distearyloxypropyl. The PEG-lipid can be one or more of PEG-DMG, PEG-dilaurylglycerol, PEG-dipalmitoylglycerol, PEG- disterylg lycerol, PEG-dilaurylglycamide, PEG-dimyristylglycamide, PEG- dipalmitoylglycamide, PEG- disterylglycamide, PEG-cholesterol (l-[8'-(Cholest-5-en-3[beta]- oxy)carboxamido-3',6'-dioxaoctanyl] carbamoyl-[omega]-methyl-poly(ethylene glycol), PEG- DMB (3,4-Ditetradecoxylbenzyl- [omega]- methyl-poly(ethylene glycol) ether), and 1 ,2- dimyristoyl-sn-glycero-3-phosphoethanolamine-N- [methoxy(polyethylene glycol)-2000]. In some embodiments, the PEG-lipid comprises PEG-DMG, 1 ,2- dimyristoyl-sn-glycero-3-phosphoethanolamine-N-[methoxy(polyethylene glycol)-2000]. In some embodiments, the PEG-lipid comprises a structure selected from:

In some embodiments, lipids conjugated with a molecule other than a PEG can also be used in place of PEG-lipid. For example, polyoxazoline (POZ)-lipid conjugates, polyamide-lipid conjugates (such as ATTA-lipid conjugates), and cationic-polymer lipid (GPL) conjugates can be used in place of or in addition to the PEG-lipid.

Exemplary conjugated lipids, i.e. , PEG-lipids, (POZ)-lipid conjugates, ATTA-lipid conjugates and cationic polymer-lipids are described in the PCT and LIS patent applications listed in Table 2 of WO2019051289A9, the contents of all of which are incorporated herein by reference in their entirety.

In some embodiments, the PEG or the conjugated lipid can comprise 0-20% (mol) of the total lipid present in the lipid nanoparticle. In some embodiments, PEG or the conjugated lipid content is 0.5- 10% or 2-5% (mol) of the total lipid present in the lipid nanoparticle. Molar ratios of the ionizable lipid, non-cationic-lipid, sterol, and PEG/conjugated lipid can be varied as needed. For example, the lipid particle can comprise 30-70% ionizable lipid by mole or by total weight of the composition, 0-60% cholesterol by mole or by total weight of the composition, 0-30% non-cationic-lipid by mole or by total weight of the composition and 1 -10% conjugated lipid by mole or by total weight of the composition. Preferably, the composition comprises 30-40% ionizable lipid by mole or by total weight of the composition, 40-50% cholesterol by mole or by total weight of the composition, and 10- 20% non- cationic-lipid by mole or by total weight of the composition. In some other embodiments, the composition is 50-75% ionizable lipid by mole or by total weight of the composition, 20-40% cholesterol by mole or by total weight of the composition, and 5 to 10% non-cationic-lipid, by mole or by total weight of the composition and 1 -10% conjugated lipid by mole or by total weight of the composition. The composition may contain 60-70% ionizable lipid by mole or by total weight of the composition, 25-35% cholesterol by mole or by total weight of the composition, and 5-10% non-cationic-lipid by mole or by total weight of the composition. The composition may also contain up to 90% ionizable lipid by mole or by total weight of the composition and 2 to 15% non-cationic lipid by mole or by total weight of the composition. The formulation may also be a lipid nanoparticle formulation, for example comprising 8- 30% ionizable lipid by mole or by total weight of the composition, 5-30% non- cationic lipid by mole or by total weight of the composition, and 0-20% cholesterol by mole or by total weight of the composition; 4-25% ionizable lipid by mole or by total weight of the composition, 4-25% non-cationic lipid by mole or by total weight of the composition, 2 to 25% cholesterol by mole or by total weight of the composition, 10 to 35% conjugate lipid by mole or by total weight of the composition, and 5% cholesterol by mole or by total weight of the composition; or 2-30% ionizable lipid by mole or by total weight of the composition, 2-30% non-cationic lipid by mole or by total weight of the composition, 1 to 15% cholesterol by mole or by total weight of the composition, 2 to 35% conjugate lipid by mole or by total weight of the composition, and 1 -20% cholesterol by mole or by total weight of the composition; or even up to 90% ionizable lipid by mole or by total weight of the composition and 2-10% non-cationic lipids by mole or by total weight of the composition, or even 100% cationic lipid by mole or by total weight of the composition. In some embodiments, the lipid particle formulation comprises ionizable lipid, phospholipid, cholesterol and a PEG-ylated lipid in a molar ratio of 50: 10:38.5: 1 .5. In some other embodiments, the lipid particle formulation comprises ionizable lipid, cholesterol and a PEG-ylated lipid in a molar ratio of 60:38.5:1 .5.

In some embodiments, the lipid particle comprises ionizable lipid, non-cationic lipid (e.g. phospholipid), a sterol (e.g., cholesterol) and a PEG-ylated lipid, where the molar ratio of lipids ranges from 20 to 70 mole percent for the ionizable lipid, with a target of 40-60, the mole percent of noncationic lipid ranges from 0 to 30, with a target of 0 to 15, the mole percent of sterol ranges from 20 to 70, with a target of 30 to 50, and the mole percent of PEG-ylated lipid ranges from 1 to 6, with a target of 2 to 5.

In some embodiments, the lipid particle comprises ionizable lipid I non-cationic- lipid I sterol I conjugated lipid at a molar ratio of 50:10:38.5:1 .5.

In an aspect, the disclosure provides a lipid nanoparticle formulation comprising phospholipids, lecithin, phosphatidylcholine and phosphatidylethanolamine.

In some embodiments, one or more additional compounds can also be included. Those compounds can be administered separately, or the additional compounds can be included in the lipid nanoparticles of the invention. In other words, the lipid nanoparticles can contain other compounds in addition to the nucleic acid or at least a second nucleic acid, different than the first. Without limitations, other additional compounds can be selected from the group consisting of small or large organic or inorganic molecules, monosaccharides, disaccharides, trisaccharides, oligosaccharides, polysaccharides, peptides, proteins, peptide analogs and derivatives thereof, peptidomimetics, nucleic acids, nucleic acid analogs and derivatives, an extract made from biological materials, or any combinations thereof.

In some embodiments, LNPs are directed to specific tissues by the addition of LNP targeting domains. For example, biological ligands may be displayed on the surface of LNPs to enhance interaction with cells displaying cognate receptors, thus driving association with and cargo delivery to tissues wherein cells express the receptor. In some embodiments, the biological ligand may be a ligand that drives delivery to the liver, e.g., LNPs that display GalNAc result in delivery of nucleic acid cargo to hepatocytes that display asialoglycoprotein receptor (ASGPR). The work of Akinc et al. Mol Ther 18(7):1357-1364 (2010) teaches the conjugation of a trivalent GalNAc ligand to a PEG-lipid (GalNAc- PEG-DSG) to yield LNPs dependent on ASGPR for observable LNP cargo effect (see, e.g., FIG. 6 of Akinc et al. 2010, supra). Other ligand-displaying LNP formulations, e.g., incorporating folate, transferrin, or antibodies, are discussed in WO2017223135, which is incorporated herein by reference in its entirety, in addition to the references used therein, namely Kolhatkar et al., Curr Drug Discov Technol. 2011 8:197-206; Musacchio and Torchilin, Front Biosci. 2011 16:1388-1412; Yu et al., Mol Membr Biol. 2010 27:286-298; Patil et al., Crit Rev Ther Drug Carrier Syst. 2008 25:1 -61 ; Benoit et al., Biomacromolecules. 2011 12:2708-2714; Zhao et al., Expert Opin Drug Deliv. 2008 5:309-319; Akinc et al., Mol Ther. 2010 18:1357-1364; Srinivasan et al., Methods Mol Biol. 2012 820:105-116; Ben-Arie et al., Methods Mol Biol. 2012 757:497-507; Peer 2010 J Control Release. 20:63-68; Peer et al., Proc Natl Acad Sci U S A. 2007 104:4095-4100; Kim et al., Methods Mol Biol. 2011 721 :339-353; Subramanya et al., Mol Ther. 2010 18:2028-2037; Song et al., Nat Biotechnol. 2005 23:709-717; Peer et al., Science. 2008 319:627-630; and Peer and Lieberman, Gene Ther. 2011 18:1127-1133.

In some embodiments, LNPs are selected for tissue-specific activity by the addition of a Selective ORgan Targeting (SORT) molecule to a formulation comprising traditional components, such as ionizable cationic lipids, amphipathic phospholipids, cholesterol and polyethylene glycol) (PEG) lipids. The teachings of Cheng et al. Nat Nanotechnol 15(4):313-320 (2020) demonstrate that the addition of a supplemental “SORT” component precisely alters the in vivo RNA delivery profile and mediates tissue-specific (e.g., lungs, liver, spleen) gene delivery and editing as a function of the percentage and biophysical property of the SORT molecule.

In some embodiments, the LNPs comprise biodegradable, ionizable lipids. In some embodiments, the LNPs comprise (9Z,l2Z)-3-((4,4-bis(octyloxy)butanoyl)oxy)-2-((((3- (diethylamino)propoxy)carbonyl)oxy)methyl)propyl octadeca-9,l2-dienoate, also called 3- ((4,4- bis(octyloxy)butanoyl)oxy)-2-((((3-(diethylamino)propoxy)carbonyl)oxy)methyl)propyl (9Z,l2Z)-octadeca- 9,12-dienoate) or another ionizable lipid. See, e.g, lipids of WO2019/067992, WO/2017/173054, WO2015/095340, and WO2014/136086, as well as references provided therein. In some embodiments, the term cationic and ionizable in the context of LNP lipids is interchangeable, e.g., wherein ionizable lipids are cationic depending on the pH.

In some embodiments, the average LNP diameter of the LNP formulation may be between 10s of nm and 100s of nm, e.g., measured by dynamic light scattering (DLS). In some embodiments, the average LNP diameter of the LNP formulation may be from about 40 nm to about 150 nm, such as about 40 nm, 45 nm, 50 nm, 55 nm, 60 nm, 65 nm, 70 nm, 75 nm, 80 nm, 85 nm, 90 nm, 95 nm, 100 nm, 105 nm, 110 nm, 115 nm, 120 nm, 125 nm, 130 nm, 135 nm, 140 nm, 145 nm, or 150 nm. In some embodiments, the average LNP diameter of the LNP formulation may be from about 50 nm to about 100 nm, from about 50 nm to about 90 nm, from about 50 nm to about 80 nm, from about 50 nm to about 70 nm, from about 50 nm to about 60 nm, from about 60 nm to about 100 nm, from about 60 nm to about 90 nm, from about 60 nm to about 80 nm, from about 60 nm to about 70 nm, from about 70 nm to about 100 nm, from about 70 nm to about 90 nm, from about 70 nm to about 80 nm, from about 80 nm to about 100 nm, from about 80 nm to about 90 nm, or from about 90 nm to about 100 nm. In some embodiments, the average LNP diameter of the LNP formulation may be from about 70 nm to about 100 nm. In a particular embodiment, the average LNP diameter of the LNP formulation may be about 80 nm. In some embodiments, the average LNP diameter of the LNP formulation may be about 100 nm. In some embodiments, the average LNP diameter of the LNP formulation ranges from about I mm to about 500 mm, from about 5 mm to about 200 mm, from about 10 mm to about 100 mm, from about 20 mm to about 80 mm, from about 25 mm to about 60 mm, from about 30 mm to about 55 mm, from about 35 mm to about 50 mm, or from about 38 mm to about 42 mm.

A LNP may, in some instances, be relatively homogenous. A polydispersity index may be used to indicate the homogeneity of a LNP, e.g., the particle size distribution of the lipid nanoparticles. A small (e.g., less than 0.3) polydispersity index generally indicates a narrow particle size distribution. A LNP may have a polydispersity index from about 0 to about 0.25, such as 0.01 , 0.02, 0.03, 0.04, 0.05, 0.06, 0.07, 0.08, 0.09, 0.10, 0.11 , 0.12, 0.13, 0.14, 0.15, 0.16, 0.17, 0.18, 0.19, 0.20, 0.21 , 0.22, 0.23, 0.24, or 0.25. In some embodiments, the polydispersity index of a LNP may be from about 0.10 to about 0.20.

The zeta potential of a LNP may be used to indicate the electrokinetic potential of the composition. In some embodiments, the zeta potential may describe the surface charge of an LNP. Lipid nanoparticles with relatively low charges, positive or negative, are generally desirable, as more highly charged species may interact undesirably with cells, tissues, and other elements in the body. In some embodiments, the zeta potential of a LNP may be from about -10 mV to about +20 mV, from about -10 mV to about +15 mV, from about -10 mV to about +10 mV, from about -10 mV to about +5 mV, from about -10 mV to about 0 mV, from about -10 mV to about -5 mV, from about -5 mV to about +20 mV, from about -5 mV to about +15 mV, from about -5 mV to about +10 mV, from about -5 mV to about +5 mV, from about -5 mV to about 0 mV, from about 0 mV to about +20 mV, from about 0 mV to about +15 mV, from about 0 mV to about +10 mV, from about 0 mV to about +5 mV, from about +5 mV to about +20 mV, from about +5 mV to about +15 mV, or from about +5 mV to about +10 mV.

The efficiency of encapsulation of a protein and/or nucleic acid, describes the amount of protein and/or nucleic acid that is encapsulated or otherwise associated with a LNP after preparation, relative to the initial amount provided. The encapsulation efficiency is desirably high (e.g., close to 100%). The encapsulation efficiency may be measured, for example, by comparing the amount of protein or nucleic acid in a solution containing the lipid nanoparticle before and after breaking up the lipid nanoparticle with one or more organic solvents or detergents. An anion exchange resin may be used to measure the amount of free protein or nucleic acid (e.g., RNA) in a solution. Fluorescence may be used to measure the amount of free protein and/or nucleic acid (e.g., RNA) in a solution. For the lipid nanoparticles described herein, the encapsulation efficiency of a protein and/or nucleic acid may be at least 50%, for example 50%, 55%, 60%, 65%, 70%, 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100%. In some embodiments, the encapsulation efficiency may be at least 80%. In some embodiments, the encapsulation efficiency may be at least 90%. In some embodiments, the encapsulation efficiency may be at least 95%.

A LNP may optionally comprise one or more coatings. In some embodiments, a LNP may be formulated in a capsule, film, or table having a coating. A capsule, film, or tablet including a composition described herein may have any useful size, tensile strength, hardness or density. Additional exemplary lipids, formulations, methods, and characterization of LNPs are taught by W02020061457, which is incorporated herein by reference in its entirety.

In some embodiments, in vitro or ex vivo cell lipofections are performed using Lipofectamine MessengerMax (Thermo Fisher) or TransIT-mRNA Transfection Reagent (Mirus Bio). In certain embodiments, LNPs are formulated using the GenVoyJLM ionizable lipid mix (Precision NanoSystems). In certain embodiments, LNPs are formulated using 2,2-dilinoleyl-4-dimethylaminoethyl- [1 ,3]-dioxolane (DLin-KC2-DMA) or dilinoleylmethyl-4-dimethylaminobutyrate (DLin-MC3-DMA or MC3), the formulation and in vivo use of which are taught in Jayaraman et al. Angew Chem Int Ed Engl 51 (34):8529-8533 (2012), incorporated herein by reference in its entirety.

LNP formulations optimized for the delivery of CRISPR-Cas systems, e.g., Cas9-gRNA RNP, gRNA, Cas9 mRNA, are described in WO2019067992 and WO2019067910, both incorporated by reference.

Additional specific LNP formulations useful for delivery of nucleic acids are described in US8158601 and US8168775, both incorporated by reference, which include formulations used in patisiran, sold under the name ONPATTRO.

Exemplary dosing of LNPs comprising the RNA compositions described herein may include about 0.1 , 0.25, 0.3, 0.5, 1 , 2, 3, 4, 5, 6, 8, 10, or 100 mg/kg (RNA). Exemplary dosing of AAV comprising a nucleic acid encoding one or more components of the system may include an MOI of about 1011 , 1012, 1013, and 1014 vg/kg.

In some embodiments, the invention includes a lipid nanoparticle (LNP) comprising the ANDbody polypeptide (or RNA encoding the same), nucleic acid molecule, or DNA encoding an ANDbody described herein. In embodiments, the LNP comprises a cationic lipid. In some embodiments, the LNP further comprises one or more neutral lipid, e.g., DSPC, DPPC, DMPC, DOPC, POPC, DOPE, SM, a steroid, e.g., cholesterol, and/or one or more polymer conjugated lipid, e.g., a pegylated lipid, e.g., PEG-DAG, PEG-PE, PEG-S-DAG, PEG-cer or a PEG dialkyoxypropylcarbamate. In some embodiments, the cationic lipid of the LNP has a structure according to:

For a review of LNP, see also, e.g., Li et al. 2017, Nanomaterials 7, 122; doi:10.3390/nano7060122.

Other Carriers

Viral vectors

The compositions described herein (e.g., polypeptide or RNA ANDbody compositions), can be delivered by a viral vector (e.g., a viral vector expressing an RNA). A viral vector may be administered to a cell or to a subject (e.g., a human subject or non-human animal). A viral vector may be locally or systemically administered.

Examples of viral vectors include a retrovirus (e.g., Retroviridae family viral vector), adenovirus (e.g., Ad5, Ad26, Ad34, Ad35, and Ad48), parvovirus (e.g., adeno-associated viruses), coronavirus, negative strand RNA viruses such as orthomyxovirus (e.g., influenza virus), rhabdovirus (e.g., rabies and vesicular stomatitis virus), paramyxovirus (e.g., measles and Sendai), positive strand RNA viruses, such as picornavirus and alphavirus, and double stranded DNA viruses including adenovirus, herpesvirus (e.g., Herpes Simplex virus types 1 and 2, Epstein-Barr virus, cytomegalovirus, replication deficient herpes virus), and poxvirus (e.g., vaccinia, modified vaccinia Ankara (MVA), fowlpox and canarypox). Other viruses include Norwalk virus, togavirus, flavivirus, reoviruses, papovavirus, hepadnavirus, human papilloma virus, human foamy virus, and hepatitis virus, for example. Examples of retroviruses include: avian leukosis-sarcoma, avian C-type viruses, mammalian C-type, B-type viruses, D-type viruses, oncoretroviruses, HTLV-BLV group, lentivirus, alpharetrovirus, gammaretrovirus, spumavirus (Coffin, J. M., Retroviridae: The viruses and their replication, Virology (Third Edition) Lippincott-Raven, Philadelphia, 1996). Other examples include murine leukemia viruses, murine sarcoma viruses, mouse mammary tumor virus, bovine leukemia virus, feline leukemia virus, feline sarcoma virus, avian leukemia virus, human T-cell leukemia virus, baboon endogenous virus, Gibbon ape leukemia virus, Mason Pfizer monkey virus, simian immunodeficiency virus, simian sarcoma virus, Rous sarcoma virus and lentiviruses. Other examples of vectors are described, for example, in US Patent No. 5,801 ,030, the teachings of which are incorporated herein by reference.

Anellovirus vectors can also be used for delivering an ANDbody composition described herein. Anellovectors are known in the art and described, e.g., in W02020123773, WO2020123816, WO2018232017, and W02020123773. In certain embodiments, an anellovector composition comprises a genomic element that comprises a promoter operably linked to a nucleic acid sequence encoding an ANDbody described herein, the genetic element encapsulated by a proteinaceous exterior comprising an Anellovirus ORF1 , e.g., an anellovirus capsid protein.

Cell and vesicle-based carriers

A composition described herein {e.g., polypeptide or RNA ANDbody compositions), described herein can be administered to a cell in a cell, vesicle or other membrane-based carrier. In one embodiment, the compositions and systems described herein can be formulated in liposomes or other similar vesicles. Liposomes are spherical vesicle structures composed of a uni- or multilamellar lipid bilayer surrounding internal aqueous compartments and a relatively impermeable outer lipophilic phospholipid bilayer. Liposomes may be anionic, neutral or cationic. Liposomes are biocompatible, nontoxic, can deliver both hydrophilic and lipophilic drug molecules, protect their cargo from degradation by plasma enzymes, and transport their load across biological membranes and the blood brain barrier (BBB) (see, e.g., Spuch and Navarro, Journal of Drug Delivery, vol. 2011 , Article ID 469679, 12 pages, 2011. doi :10.1155/2011/469679 for review). Vesicles can be made from several different types of lipids; however, phospholipids are most commonly used to generate liposomes as drug carriers. Methods for preparation of multilamellar vesicle lipids are known in the art (see for example U.S. Pat. No. 6,693,086, the teachings of which relating to multilamellar vesicle lipid preparation are incorporated herein by reference). Although vesicle formation can be spontaneous when a lipid film is mixed with an aqueous solution, it can also be expedited by applying force in the form of shaking by using a homogenizer, sonicator, or an extrusion apparatus (see, e.g., Spuch and Navarro, Journal of Drug Delivery, vol. 2011 , Article ID 469679, 12 pages, 2011 . doi:10.1155/2011/469679 for review). Extruded lipids can be prepared by extruding through filters of decreasing size, as described in Templeton et al., Nature Biotech, 15:647-652, 1997, the teachings of which relating to extruded lipid preparation are incorporated herein by reference.

Exosomes can also be used as drug delivery vehicles for the compositions and systems described herein. For a review, see Ha et al. July 2016. Acta Pharmaceutica Sinica B. Volume 6, Issue 4, Pages 287-296; https://doi.Org/10.1016/j.apsb.2016.02.001.

Ex vivo differentiated red blood cells can also be used as a carrier for an agent (e.g., an inhibitor) described herein, e.g., an antibody or a nucleic acid described herein. See, e.g., WO2015073587; WO2017123646; WO2017123644; WO2018102740; wO2016183482;

WO2015153102; WO2018151829; WO2018009838; Shi et al. 2014. Proc Natl Acad Sci USA. 111 (28): 10131-10136; US Patent 9,644,180; Huang et al. 2017. Nature Communications 8: 423; Shi et al. 2014. Proc Natl Acad Sci USA. 111 (28): 10131-10136.

Fusosome compositions, e.g., as described in WO2018208728, can also be used as carriers to deliver the [agent] or preparation described herein.

Plant nanovesicles and plant messenger packs (PMPs), e.g., as described in WO2011097480, WO2013070324, WO2017004526, or W02020041784 can also be used as carriers to deliver the compositions described herein.

Without further elaboration, it is believed that one skilled in the art can, based on the above description, utilize the present invention to its fullest extent. The following specific embodiments are, therefore, to be construed as merely illustrative, and not limitative of the remainder of the disclosure in any way whatsoever. All publications and sections thereof cited herein are herein incorporated by reference for the purposes or subject matter referenced herein. EXAMPLES

The invention will be further illustrated in the following non-limiting examples.

TABLE OF CONTENTS

Example 1 : Design of ANDbody binding Chondroadherin (CHAD) address and Sclerostin (SOST) target a. Vaccination to create anti- CHAD antibodies

Antibodies against the human chondroadherin (CHAD) extracellular domain are created by immunization. The extracellular domain of human CHAD (NCBI protein accession NP_001258, positions A22-H359) (huCHAD) fused to the Fc region of human lgG1 (UniProt ID P01857, positions P100-K330) is expressed in HEK293F cells. Briefly, DNA sequences are codon optimized for mammalian expression and placed in an expression vector. Proteins are transiently transfected into cells (e.g., HEK293 cells) and purified using Protein A affinity purification as per routine methods. 50 pg of the huCHAD-Fc fusion protein is used to immunize female BALB/c mice by intraperitoneal (i.p.) injection in Complete Freund's Adjuvant or Incomplete Freund's Adjuvant (CFA/IFA adjuvant). Subsequently, hybridomas are generated (Listek et al., Scientific Reports, 10: 1664, 2020). Clones are initially screened for IgG reactivity specific for the huCHAD-Fc fusion protein used for immunization in an ELISA format, followed by flow cytometry studies using cells stably (CHO) or transiently (HEK293F) transfected with full-length huCHAD. Anti-CHAD hybridoma clones are next evaluated based on murine cross-reactivity. Flow cytometry studies are done using cells stably (CHO) or transiently (HEK293F) transfected with full-length mouse CHAD (mCHAD), and clones are selected that bind to mCHAD. Positive clones expressing anti-CHAD mAbs cross-reactive between human and mouse are then further purified by limited dilution cloning. The hybridomas are grown in DMEM/2% ultra-low IgG serum and the mAbs are purified by protein G chromatography according to the routine methods. b. Selecting for inert anti-CHAD antibodies

Address target binding sites are designed to not significantly modulate the function of the address target. Accordingly, anti-CHAD hybridoma clones produced as described above are further evaluated based on their inability to block chondrocyte alpha2 betal integrin-dependent adhesion. Recombinant human CHAD protein (R&D Systems Catalog # 8218-CH-050) is used to coat tissue culture 24-well plates (5 pg/ml in denaturing conditions using 4M guanidine hydrochloride). After overnight incubation, plates are washed with phosphate-buffered saline (PBS) and unspecific binding is blocked using PBS with 1% bovine serum albumin (BSA) for 1 hour at room temperature.

The murine chondrocyte cell line ATCD5 (Sigma Catalog # 99072806) is evaluated for its capacity to bind to the wells in the presence of 10 fM up to 10 pM of anti-CHAD antibody clones from the hybridomas (one condition per concentration and per clone). 80,000 ATCD5 cells are plated in tissue culture medium in presence of anti-CHAD antibodies for 1 hour and unbound cells are washed with PBS. The number of adherent cells is calculated using the VYBRANT™ Cell Adhesion Assay Kit (ThermoFisher, Catalog # V13181 ) according to the manufacturer’s instructions. Isotype unrelated (negative) human IgG 1 antibodies are used as controls in this assay. c. Identifying anti-CHAD antibodies that bind specifically to mouse bone

Address binding targets of the present technology, such as the exemplary CHAD address binding target, are designed to bind specifically to a target tissue or cell type, while minimizing binding to other tissues. Accordingly, anti-CHAD hybridoma clones produced as described above are further evaluated for specificity to bone. This is achieved in two steps, both using immunohistochemistry (IHC). Firstly, binding to bone matrix is determined by incubating decalcified human bone tissue sections mounted onto glass slides with anti-CHAD antibodies. Antibody binding to bone tissue is detected by the incubation of a horse radish peroxidase (HRP) conjugated secondary antibody reactive to mouse antibodies. Location and intensity of binding is determined by the addition of the HRP substrate 3,3’Diaminobenzidine (DAB), which yields a brown color at the site of primary antibody binding, proportional to the abundance of deposited antibody. Nuclei are counterstained with hematoxylin to yield a blue color. Secondly, the specificity of anti-CHAD antibodies is determined by performing IHC (as above on bone tissue) on fresh frozen human tissue microarray (TMA) sections mounted onto glass slides. The TMA comprise non-bone tissues representative of the main tissue constituents of the rest of the body (e.g. spleen, heart, kidney (medulla and cortex), lung (upper and lower airway), skin, liver, pancreas, large intestine (ascending, descending and transverse), small intestine (duodenum, jejunum and ileum), stomach, heart, skeletal muscle and brain). Specificity of the antibody is determined by the relative reactivity with bone compared to cells in any of the cells on the TMA (which does not contain bone). d. Vaccination to create anti-SOST antibodies

Sclerostin (SOST) protein (huSOST) is an exemplary effector target. Human SOST protein (NCBI protein accession NP_079513, positions Q24-Y213) fused to the Fc region of human lgG1 are created by immunization using a process similar to the process described above for CHAD. After immunization and hybridoma generation, clones are screened as described above, but for binding to full length human and mouse SOST (instead of CHAD). Positive clones expressing anti-SOST mAbs cross-reactive between human and mouse are then further purified by limited dilution cloning. The hybridomas are grown in DMEM/2% ultra-low IgG serum and the mAbs are purified by protein G chromatography. e. Selecting for anti-SOST antibodies with a wide range of inhibitory concentrations and in vitro functions

Effector target binding antibodies of the present technology, such as SOST, are designed to have little or no pharmacological effect unless they localize at a target binding site, such as CHAD protein-positive bone matrix. Accordingly, the effector target binding antibody is selected such that binding to a localized target is necessary to achieve a sustained pharmacologically active concentration.

Anti-SOST antibodies are selected with a range of inhibitory concentrations (IC50), such as < 1 nM to 1 pM, when evaluated for ability to block SOST activity in vitro. SOST blocking activity is measured in beta-catenin dependent Wnt reporter assays. For these assays, a HEK293 cell line containing a T cell factor/lymphoid enhancer factor (TCF/LEF) transgene reporter that drives luciferase expression to measure activation of the Wnt signaling pathway (BPS Bioscience Catalog # 60501 ) is used. 35,000 cells are plated overnight in 96-well plates in 100 pL of medium containing 10mM lithium chloride (LiCI). After 18 hours, cells are treated with soluble recombinant murine Wnt1 (37 ng/mL) (R&D Systems Catalog # 9765-WN-010) or human Wnt3a (111.1 ng/mL) (R&D Systems Catalog # 5036-WN) together with SOST (1 pg/mL) and serial dilutions of the anti-SOST antibody (10 pg/mL and 1 :3 dilutions). After 5 hours at 37°C, cells are lysed in 100 pL of Luciferase Reagent Buffer (Component A) containing 1 pL of Luciferase Reagent Substrate (Component B) (BPS Biosciences Catalog # 60690) that is added to the wells. Luminescence is evaluated using an SPECTRAMAX® i3x Multi-Mode Microplate Reader. Wnt activation of reporter cells leads to luciferase activity that is blocked by SOST. Blocking of SOST activity by antibodies is evaluated as an increase in luciferase signal.

Functional in vitro assays are performed using MC3T3 osteoblastic cells (ATCC Catalog # CRL-2593). For evaluation of mineralization, 10,000 MC3T3 cells are plated in 24-well plates and allowed to reach confluency. Mineralization medium containing 50 pg/mL of Vitamin C, 10 mM betaglycerolphosphate, and 20 ng/mL of bone morphogenetic protein-2 (BMP-2) is added. 400 ng/mL of SOST and serial dilutions of the anti-SOST antibody (10 pg/mL and 1 :3 dilutions) are added to the wells and medium with the corresponding additives is replaced every 2-3 days. After 8 days, cells are washed and fixed in 10% formalin for 10 minutes, washed 3 times with water, and incubated with 500 pl of 40mM Alizarin solution (Millipore Sigma catalog #TMS-008-C). After 15 minutes, cells are washed 3 times with water. For calcium quantification, 500 pl of 10% acetic acid (v/v) is added to the cells for 30 minutes at room temperature (RT). Cells are scraped and transferred to Eppendorf tubes, vortexed for 30 seconds, and incubated for 15 minutes at 85°C. Samples are centrifuged for 15 minutes at 20,000 x g and 250 pl of supernatant is transferred to another Eppendorf tube. 100 pl of 10% NH4OH (v/v) are added to the samples and mixed. The absorbance is measured at 405 nm using a SPECTRAMAX® i3x Multi-Mode Microplate Reader. f. Reducing the affinity of an anti-SOST antibody

Effector binding antibodies may need to have their binding affinity reduced to meet the requirements of design for an ANDbody effector target binding domain. In exemplary anti-SOST antibodies, the affinity to SOST may be too strong compared with the address target binder. In this case, dematuration prior to the generation of the corresponding ANDbody is performed. This provides a repertoire of antibodies that favor targeting to the bone address target where the pharmacodynamic effect is locally targeted (such as KD > 100 nM). Affinity dematuration is carried out by alanine scanning mutagenesis in selected residues of individual complementarity-determining regions (CDR) of the variable light chain (VL) of the anti-SOST antibody (Phasel ).

To evaluate binding affinity of anti-SOST antibodies and address antibodies, the dissociation constant (KD), on-rate (k_On), and off-rate (k_Off) of the antibody binding to the antigen is measured using biolayer interferometry (GatorPrime). Polypropylene 96-well black, F-bottom plates (RATIOLAB®) are utilized for this experiment. Briefly, pre-equilibrated anti-human IgG Fc capture biosensor tips (Gator Bio) are baselined in 1 x K buffer (Gator Bio), and antibodies are loaded onto the anti-human IgG Fc capture tips at a concentration of 2 pg/mL. An additional baseline step is performed in 1 x K buffer followed by an association step using varying concentrations of the antigen (0-300 nM). For analysis and determination of the KD for each antigen/antibody pair, a global fit is performed using all concentrations of antigen for which the association response is greater than 0.1 and the dissociation results in a measurable off-rate. A 1 :1 model is utilized for all curve fitting. The data are analyzed using GATOR™ GatorOne analysis software (Gator Bio).

If changes in individual VL CDR residues do not result in affinities equal to or less than the 100 nM range, multiple combinatorial alanine substitutions are generated, and newly synthesized constructs are evaluated for changes in affinity (Phase2). Lack of affinity reduction after Phases 1 and 2 is followed by alanine scanning mutagenesis in the variable heavy (VH) sequence (Phase 3) and evaluation of antibody affinity to SOST by BLI.

Example 2: Production and characterization of ANDbodies binding CHAD address and SOST target a. Production of ANDbodies

DNA sequences encoding 10 anti-CHAD antibodies and 10 anti-SOST antibodies ranging in ICsofrom < 1 nM up to 5 pM are cloned using IN-FUSION® HD Cloning (Takara Bio, Catalog # 638911 ) into a human lgG1 framework with single matching point mutations in the CH3 domain Fc region according to the ‘Controlled Fab-Arm Exchange’ (cFAE) method (Labrijn et al., Nature Protocols, 9(10): 2450-2463, 2014). After separately expressing antibodies from transient HEK293 expressions and purifying each antibody using protein A affinity resin, parental antibodies (combinations of one anti-CHAD antibody with one anti-SOST antibody) are made into anti- CHAD/SOST ANDbodies according to the cFAE method. Briefly, parental antibodies are mixed under permissive redox conditions to enable recombination of half-molecules. Subsequently, the reductant is removed to allow for reoxidation of interchain disulfide bonds. Lastly, exchange efficiency is quantified using chromatography-based or mass spectrometry-based methods. Around 100 variant ANDbodies of anti-CHAD/SOST are made. b. Simultaneous binding to CHAD and SOST in vitro

To identify ANDbody variants that meet desired effector target and address target binding affinity criteria, a dual binding assay to two different antigens is performed using a BLI assay as described above with the following modifications. Briefly, the ANDbody candidates are loaded onto the sensor tips (anti-human IgG Fc) pre-equilibrated with K buffer followed by a baseline step. The loaded tip is dipped into the well containing purified soluble CHAD (first association step) and then into the well of the purified soluble SOST (second association step), followed by a dissociation step. BLI-based affinity measurements are carried out as described above. ANDbody variants with higher affinity for CHAD than SOST as well as variants with no affinity differences or higher affinity for SOST are used in subsequent in vitro and in vivo experiments. c. In vitro assays for Sclerostin-blocking activity in ANDbodies

ANDbodies of the present technology must retain functional properties of the parental effector target binding antibody, and the binding affinity must be such that the pharmacological effect can be localized by binding to the address target. In the context of anti-CHAD/SOST ANDbodies, the inhibitory potency (IC50) for SOST is determined as above by measuring relief of inhibition by SOST in the Wnt signaling assay using TCF/LEF reporter cells and by measuring relief of inhibition of mineralization in MC3T3 osteoblastic cells. d. Tissue-specific binding of anti-CHAD/SOST ANDbodies

ANDbodies of the present technology retain tissue and/or cell binding properties similar to their parental localizer (address target binding) moiety. To identify ANDbody variants that meet this criterion, molecules with a range of affinities (such as anti-CHAD/SOST variants described above) are evaluated for binding to murine bone by IHC, tissues on a murine tissue microarray (as described above, substituting murine tissues for human tissues), and anti-human IgG 1 secondary antibody for anti-mouse secondary antibody.

Example 3: Design of ANDbody binding Interferon-induced Transmembrane Protein 5 (IFITM5) address and SOST target a. Vaccination to create anti-IFITM5 antibodies

Antibodies binding to interferon-induced Transmembrane Protein 5 (IFITM5) are produced as described above for CHAD, except that the extracellular domain of human IFITM5 (NCBI protein accession NP 001020466, positions M1 -H36) is fused to human lgG1 and used for immunization. The human and equivalent mouse protein sequences are used for identification of anti-IFITM5 antibodies. b. Selecting for inert anti-IFITM5 antibodies

Anti-IFITM5 hybridoma clones produced as described above are further evaluated based on their inability to block mineralization in osteoblastic cell lines. For this, MC3T3 cells are subjected to a mineralization assay, as described above, in the presence of 1 , 2, 5 or 10 pg/mL of selected anti- IFITM5 antibodies or an antibody control. Antibodies that do not interfere with the process of mineralization are selected for their ability, in vitro, to block SOST activity as measured in beta- catenin dependent Wnt reporter assays as described above.

Additionally, human osteosarcoma SaOS2 cells (ATCC Catalog # HTB-85) are transfected with human IFITM5 cDNA cloned into pCDNA3.1 using Saos-2 CELL AVALANCHE™ Transfection Reagent (EZBIOSYSTEMS™ Catalog # EZT-SAOS-1 ) according to manufacturer protocols. For this, 1 x10⁴ SaOS2 cells are plated in 96-well plates for 24 hours prior to transfection. Two days after transfection, the cells are subjected to osteogenic differentiation by supplementation of the medium with 10~⁸ M dexamethasone, 50 pg/mL vitamin C, and 10 mM p-glycerophosphate in the presence of 1 , 2, 5 or 10 ug/mL of selected anti-IFITM5 antibodies, ANDbodies, or antibody control. After three days, cells are harvested and mRNA extracted (RNeasy Micro Kit, Qiagen Catalog # 74004) and reverse-transcribed into cDNA (First Strand cDNA Synthesis, Roche, Catalog # 11483188001 ). qPCR is performed (POWERUP™ SYBR™ Green Master Mix, APPLIED BIOSYSTEMS™, Catalog # A25741 ) using an APPLIED BIOSYSTEMS™ 7500 Real-Time PCR system (APPLIED BIOSYSTEMS™). Osteogenesis-related genes evaluated include Runx2, ALP and OCN. Antibodies that do not interfere with upregulation of the aforementioned osteogenesis markers are selected for use in ANDbodies. c. Identifying anti-IFITM5 antibodies that bind specifically to mouse bone Binding strength and specificity of anti-IFITM5 antibodies is evaluated using IHC analysis of human bone tissue and tissue microarrays as described above.

Example 4: Production and characterization of ANDbodies binding IFITM5 address and SOST target a. Expressing and purifying ANDbodies as bispecific antibodies

ANDbodies comprising bispecific antibodies binding to both IFITM5 and SOST are expressed and purified as described above. b. Simultaneous binding to IFITM5 and SOST in vitro

The ability of ANDbodies to simultaneously engage with IFITM5 and SOST is determined and binding affinity for IFITM5 and SOST is measured using BLI as described above. c. In vitro assays for Sclerostin-blocking activity in ANDbodies

The ability of ANDbodies targeting IFITM5 and SOST to inhibit SOST activity is determined using TCF/LEF Wnt reporter cells and mineralization of MC3T3 osteoblastic cells as described above. d. Tissue-specific binding of anti-IFITM5/SOST ANDbodies

The ability of ANDbodies targeting IFITM5 and SOST to bind specifically to bone tissue and inhibit SOST activity is evaluated using IHC to test binding to murine bone, to tissues on a murine tissue microarray (as described above, substituting murine tissues for human tissues), and antihuman IgG 1 secondary antibody for anti-mouse secondary antibody.

Example 5: Design of ANDbody binding CHAD and Dickkopf-1 (DKK1) a. Antibodies for DKK1

Antibodies against Dickkopf-1 (DKK1 ) protein (huDKKI ) are an exemplary effector target for the present technology. Human DKK1 protein (NCBI protein accession NP_036374, positions T32- H266) fused to the Fc region of human IgG 1 are created by immunization similar to the protocol for CHAD described above. After immunization and hybridoma generation, clones are screened as described above, but for binding to full length human and mouse DKK1 (instead of CHAD). Positive clones expressing anti-DKK1 mAbs cross-reactive between human and mouse are then further purified by limited dilution cloning. The hybridomas are grown in DMEM/2% ultra-low IgG serum and the mAbs are purified by protein G chromatography. b. Selecting for anti-DKK1 antibodies with a wide range of IC50

Anti-DKK1 antibodies are selected with a range of inhibitory concentrations (IC50), such as < 1 nM to 1 pM, when evaluated for ability to block DKK1 activity in vitro. DKK1 blocking activity is measured in beta-catenin dependent Wnt reporter assays. For these assays, a HEK (human embryonic kidney) 293 cell line containing a TCF/LEF transgene reporter that drives luciferase expression to measure activation of the Wnt signaling pathway (BPS Bioscience Catalog # 60501 ) is used. 35,000 cells are plated overnight in 96-well plates in 100 pl of medium containing 10mM LiCI. After 18 hours, cells are treated with soluble recombinant murine Wnt1 (37 ng/mL) (R&D Systems Catalog # 9765-WN-010) or human Wnt3a (1 1 1 .1 ng/mL) (R&D Systems Catalog # 5036-WN) together with DKK1 (1 pg/mL) and serial dilutions of the anti-DKK1 antibody (10 pg/mL and 1 :3 dilutions). After 5 hours at 37°C, cells are lysed in 100 pl of Luciferase Reagent Buffer (Component A) containing 1 pl of Luciferase Reagent Substrate (Component B) (BPS Biosciences Catalog # 60690) that is added to the wells. Plates are gently rocked for 15 minutes at room temperature, and luminescence is evaluated using a SPECTRAMAX® i3x Multi-Mode Microplate Reader. Wnt activation of reporter cells leads to luciferase activity that is blocked by DKK1 .

Functional in vitro assays are performed using MC3T3 osteoblastic cells (ATCC Catalog # CRL-2593). For evaluation of mineralization, 10,000 MC3T3 cells are plated in 24 well plates and allowed to reach confluency. Mineralization medium containing 50 pg/mL of Vitamin C, 10mM betaglycerolphosphate, and 20 ng/mL of bone morphogenetic protein-2 (BMP-2) is added. 400ng/mL of DKK1 and serial dilutions of the anti-DKK1 antibody (10 pg/mL and 1 :3 dilutions) are added to the wells and medium with the corresponding additives is replaced every 2-3 days. After 8 days, cells are washed and fixed in 10% formalin for 10 minutes, washed 3 times with water, and incubated with 500 pl of 40mM Alizarin solution. After 15 minutes, cells are washed 3 times with water. For calcium quantification, 500 pl of 10% acetic acid (v/v) is added to the cells for 30 minutes at RT. Cells are scraped and transferred to Eppendorf tubes, vortexed for 30 seconds, and incubated for 15 minutes at 85°C. Samples are centrifuged for 15 minutes at 20,000 x g and 250 pl of supernatant is transferred to another Eppendorf tube. 100 pl of 10% NH4OH (v/v) are added to the samples and mixed. The absorbance is measured at 405 nm using a SPECTRAMAX® i3x Multi-Mode Microplate Reader. c. Expression, purification and characterization of CHAD and DKK1 ANDbodies

ANDbodies comprising bispecific antibodies binding to both CHAD and DKK1 are expressed and purified as described above for anti-CHAD/SOST ANDbodies. The ability of ANDbodies targeting CHAD and DKK1 to simultaneously engage with CHAD and DKK1 is determined and binding affinity for CHAD and DKK1 is measured using BLI as described above. The ability of ANDbodies targeting CHAD and DKK1 to inhibit DKK1 activity is determined using TCF/LEF Wnt reporter cells and mineralization of MC3T3 osteoblastic cells as described above. The ability of ANDbodies targeting CHAD and DKK1 to bind specifically to bone tissue and inhibit DKK1 activity is evaluated using IHC to test binding to murine bone, to tissues on a murine tissue microarray (as described above, substituting murine tissues for human tissues), and anti-human IgG 1 secondary antibody for antimouse secondary antibody.

Example 6: Pharmacokinetics and tissue distribution of bone-targeted ANDbodies

To analyze ANDbody distribution in vivo, the biodistribution of the exemplary ANDbodies (such as anti-CHAD/SOST, anti-IFITM5/SOST, or anti-CHAD/DKK1 ), as well as each of the parental antibodies (anti-CHAD, anti-l FITM5, anti-SOST, or anti-DKK1 used for the cFAE of the ANDbody), are quantified in female Balb/c and C57BL/6 mice.

To quantify tissue biodistribution, each antibody and ANDbody is injected individually at doses of 1 mg/kg, 3 mg/kg, and 10 mg/kg IV (tail vein). Saline (PBS) is also injected as a control at equal volume. At time points of 12 hours, 1 day, 2 days, 3 days, 7 days, and 14 days after injection, mice are euthanized using CO2 and tissues including bone matrix, heart, lung, spleen, blood, kidney, liver, and intestines are processed to homogenate. The concentration of total protein in each sample is measured using the PIERCE™ Rapid Gold BCA Protein Assay Kit (ThermoFisher Scientific, A53225). ELISA assays are used to determine the concentration of each ANDbody or antibody in homogenate containing 0.5 mg of protein.

To quantify cellular biodistribution, the ANDbody and antibodies are first individually labeled with Alexa Fluor 647 (AF647) according to the manufacturer’s instructions (ThermoFisher Scientific, A20186) using routine methods.

To determine cellular biodistribution, at time points of 12 hours, 1 day, 2 days, 3 days, 7 days, and 14 days after injection, mice are euthanized using CO2 and tissues including bone matrix, heart, lung, spleen, blood, kidney, liver, and intestines are processed into single cell suspensions according to methods previously described. Briefly, blood is collected by cardiac puncture into EDTA-treated tubes (BD Catalog # 365974), and other tissues are harvested, weighed, mechanically dissociated between frosted glass slides, and rendered into single-cell suspensions by filtration through 70-pm mesh screens (Millipore Sigma, Catalog # CLS431751 -50EA). Splenocytes, whole blood, and lung are treated with ammonium-chloride-potassium (ACK) lysing buffer (ThermoFisher Scientific, Catalog # A1049201 ). Heart is digested with collagenase and processed into single cell suspension according to previous methods (Covarrubias et al. 2019 Am J Physiol Heart Circ Physiol. 317(3):H658-H666 ). Flow cytometry is performed on immune cells using markers for CD8 T cells (CD3e+ CD8+), CD4 T cells (CD3e+ CD4+ Foxp3-), regulatory T cells (CD4+ CD25+ FOXP3+), monocytes/macrophages (CD3e- CD11 b+ CD11 c-ZIo NK1 .1 - Ly6G- SSCIo), dendritic cells (CD3e- CD11 chi), NK cells (NK1.1 + CD3e-), and NKT cells (NK1.1 + CD3e+) using routine methods. Lung cells including epithelial (CD326+CD31 -CD45-), endothelial (CD326-CD31 +CD45-), and hematopoietic lineages (CD326-CD31 -CD45+) are also analyzed.

To quantify tissue biodistribution by I VIS, the proteins (ANDbody and antibodies) are first individually labeled with NHS-5/6-FAM (ThermoFisher Scientific, Catalog # 46409) as per the manufacturer’s instructions. Mice are dosed with each ANDbody or antibody as above. At time points of 12 hours, 1 day, 2 days, 3 days, 7 days, and 14 days after injection, mice are euthanized using CO2 and tissues including bone, lung, spleen, blood, kidney, liver, and intestines are harvested, weighed, and imaged on an I VIS® Spectrum In Vivo Imaging System (Caliper Life Sciences; excitation, 500nm; emission, 540nm). Images are analyzed using the LIVING IMAGE® software (PERKINELMER®).

Example 7: In vivo effect of bone-targeted ANDbodies a. Bone-targeted antibodies are tested in a rat model of osteoporosis

Sixteen-week-old virgin female Sprague Dawley rats (Harlan) are sham operated (Sham) or ovariectomized (OVX) and left untreated for 8 weeks to allow osteopenia to develop. OVX rats are treated with control antibody or ANDbodies at a range of concentrations (such as 5, 10, 20 mg/kg, subcutaneous weekly) for 1 , 2 or 4 months. Body weights are recorded, and areal bone mineral density (BMD) is determined in vivo in anesthetized rats using dual-energy X-ray absorptiometry (DXA) methodology (Hologic FAXITRON® DXA x-ray cabinet). BMD recordings and blood samples are obtained every 3 weeks for 9 weeks until necropsy. Two weeks and 4 days before necropsy, rats are injected subcutaneously with with calcein (20 mg/kg; Sigma-Aldrich). Lumbar vertebrae, tibias, and femora are collected for further analysis including dynamic histomorphometry, microcomputed tomography, and biomechanical analysis.

For bone histomorphometry analyses, femora and lumbar vertebrae are dissected using a low-speed diamond saw, dehydrated in isopropanol, and embedded in methyl methacrylate (MMA). The cured blocks are sectioned using a microtome (Leica, Wetzlar, Germany), and 5 pm slices are examined under a fluorescent microscope to trace the calcein label. Images are analyzed using Imaged software. Static and dynamic parameters evaluated are cortical thickness, cortical thickness, endosteal and periosteal mineralizing surface/bone surface, and mineral apposition and bone formation rate.

Biochemical analyses of bone turnover at the time of necropsy may include evaluation of serum levels of osteocalcin (Rat Osteocalcin ELISA, Novus Biologicals, Catalog # NBP2-68153), P1 NP (Rat Procollagen Type 1 N-Terminal Propeptide ELISA, Novus Biologicals Catalog # NBP2- 76467), and Rat TRACP-5b (Tartrate Resistant Acid Phosphatase 5b) ELISA, Elabscience, Catalog # E-EL-R0939).

For pCT analysis, lumbar vertebrae and femoral neck and diaphysis are examined using a desktop pCT scanner (GE explore Locus Micro CT Scanner) and regions of interest are analyzed using the software’s algorithm. b. Bone-targeted antibodies are tested in a model of serum-transfer arthritis

Arthritogenic serum is harvested from 10-week-old arthritic K/BxN mice. Arthritis is induced in 12-week-old male wild-type mice with intravenous injection of 150 pL of arthritogenic serum on days 0, 2, and 6 together with ANDbodies at various concentrations (such as 2, 10 and 20 mg/kg subcutaneous). Mice are sacrificed on day 14 and forepaws are fixed in 70% ethanol for imaging. For histopathological analysis, knees and ankles are fixed in 4% paraformaldehyde, decalcified in 15% EDTA, and paraffin embedded. 4 pm knee or ankle sections are obtained for standard histopathological analysis. H&E images of various sites of periosteal bone formation at the knee joint and ankle are captured and average area for each site is calculated. This value is multiplied by the distance through the total sections obtained to calculate the total volume per site. Total periosteal bone volume for each knee or ankle is determined by totaling the volumes of bone formation from all sites. Periosteal bone formation was also quantitated using micro-computed tomographic imaging as described above.

Example 8: Production and characterization of ANDbody constructs binding DMP1 address a. Design, expression, and purification of bispecific ANDbodies for conjugation To produce dentin matrix acidic phosphoprotein 1 (DMPI )-binding address arms for later conjugation into a bispecific structure (e.g., an ANDbody construct), EXPI293F™ (Gibco) cells were transiently transfected with a 1 :2 mass ratio of heavy chain: light chain plasmid encoding an address anti-DMP1 Controlled Fab-Arm Exchange (cFAE) F405L monoclonal antibody (mAb) using polyethylenimine (PEI) and maintained according to the manufacturer’s instructions (37°C, 8% CO2 on a shaking platform). Cultures were fed 4-24 hours post-transfection to a final concentration of 5% v/v Gibco Feed B, 1 % v/v L-alanyl-glutamine, and 4 mM valproic acid. After 4-7 days, supernatants were harvested by 0.22 pm filtration. Filtered supernatants were purified via Protein A affinity chromatography (Cytiva 5 mL MABSELECT™ PrismA column) equilibrated in PBS pH 7.4 running buffer and 0.1 M Sodium Citrate pH 3.0 elution buffer. Eluted protein was immediately neutralized with 10% v/v of 1 M sodium acetate, pH 6.0, or 10% v/v of 1 M Tris, pH 8.0. Protein was further purified via Size Exclusion Chromatography equilibrated in PBS pH 7.4.

To produce SOST-binding effector arms of ANDbody constructs for later conjugation into a bispecific structure (e.g., an ANDbody construct), effector anti-SOST cFAE K409R FC star (H435R, Y436F) mAb constructs were generated to eliminate binding to protein A. EXPI293F™ cells were transiently transfected with a 1 :2 mass ratio of heavy chain: light chain plasmid using PEI and maintained according to the manufacturer’s instructions (37°C, 8% CO2 on a shaking platform). Cultures were fed 4-24 hours post-transfection to a final concentration of 5% v/v Gibco Feed B, 1 % v/v L-alanyl-glutamine, and 4 mM valproic acid. After 4-7 days, supernatants were harvested by 0.22 pm filtration. Filtered supernatants were purified via Protein G affinity chromatography (Cytiva 5 mL HITRAP® Protein G HP column) in PBS 7.4 running buffer and 0.1 M glycine pH 2.7 elution buffer. Eluted protein was immediately neutralized with 10% v/v of 1 M Tris pH 7.5. Protein was buffer exchanged into PBS for conjugation.

Conjugation of the anti-DMP1 and anti-SOST mAbs into bispecific ANDbodies was performed by mixing the above-described products in equimolar ratios in PBS pH 7.4 with 75 mM 2-MEA. The solutions were incubated at 31 °C for 5 hours, before being buffer exchanged into PBS to remove all 2-MEA. Following removal of 2-MEA, the solutions were incubated at 4°C for 16 hours. Bispecific ANDbodies were purified from the solutions using gradient elution of 40 mM Sodium Acetate pH 6, 500 mM NaCI to 100% 40 mM Sodium Acetate pH 3, 500 mM NaCI to separate properly conjugated bispecific ANDbodies from parental mAbs. b. Validation of binding to targets

Binding kinetics studies were performed on a BioLayer Interferometry (BLI)-based GATOR® Plus biosensor at 25°C in buffer containing 10mM HEPES, 150mM NaCI, 1 mg/ml BSA,0.04% NaN3, 0.05% Tween20, pH 7.4 (HBS-BNT). ANDbody molecules (e.g., conjugated or unconjugated ANDbodies) were captured by submerging anti-human Fc BLI biosensors (GatorBio - hFC) in wells containing 2 pg/mL of antibodies for 90 seconds. The biosensors were then submerged in wells containing different concentrations of the relevant antigen over a range of values for 120 seconds, followed by a 300-second dissociation in HBS-BNT buffer. All the sensors were washed in HBS-BNT buffer in between each step.

To compensate for any drift in baseline, ANDbody-captured biosensors were submerged in wells containing HBS-BNT buffer and the observed raw data were subtracted from the raw data obtained from antigen binding. This single reference subtracted data was later globally fit using the Langmuir 1 :1 binding model and the binding kinetic parameters were determined.

Example 9: Design of an ANDbody binding DMP1 address and SOST target a. Vaccination to create anti-DMP1 antibodies

Antibodies binding to dentin matrix acidic phosphoprotein 1 (DMP1 ) are produced as described for CHAD in Example 1 , except that the full-length sequence of human DMP1 (protein accession NP_004398.1 ) is fused to human lgG1 and used for immunization. The human and equivalent mouse protein sequences are used for identification of anti-DMP1 antibodies. b. Selecting for inert anti-DMP1 antibodies

Anti-DMP1 hybridoma clones produced as described above are further evaluated based on their inability to block mineralization in MC3T3 osteoblastic cells, as described above (see, e.g., Example 3(b)).

Anti-DMP1 antibodies that do not interfere with mineralization are selected for use in ANDbody generation. c. Identifying anti-DMP1 antibodies that bind specifically to mouse bone

Binding strength and specificity of anti-DMP1 antibodies is evaluated using IHC analysis of human bone tissue and tissue microarrays as described above (see, e.g., Example 1 (c)). d. Anti-SOST targeting arm

The anti-DMP1 antibodies are paired with an anti-SOST targeting arm as described in Example 1 (d), above, to generate ANDbodies binding a DMP1 address and a SOST target.

Example 10: Validation of the bone extracellular matrix address DMP1

An immunohistochemistry approach using a peroxidase-based chromogen detection system was utilized to assess DMP1 as a bone address, as well as to characterize the targeting specificity of an anti-DMP1 antibody (Fig. 5).

The anti-DMP1 murine lgG1 k monoclonal antibody Ab349 was tested for binding to bone tissue using 5-micron (pm) formalin-fixed paraffin-embedded (FFPE) decalcified mouse femur sections on standard histological microscope slides. After STANDARD deparaffinization and rehydration, samples were submerged in excess Tris EDTA solution at pH 9 (Vector Laboratories #H- 3301 -250) and incubated in a 37°C water bath overnight, serving as a gentle modification to typical heat-induced epitope retrieval (HIER). The next day, samples were treated with a commercial blocking reagent against endogenous tissue peroxidases (BLOXALL®, Vector Laboratories #SP- 6000-100). To prevent nonspecific binding of the commercial horseradish peroxidase (HRP) polymer- conjugated anti-murine IgG detection reagent to endogenous murine IgGs, the tissue sections were subsequently treated with an additional blocking reagent (Blocking+Detection reagents: Mouse on Mouse Polymer IHC Kit, Abcam#ab269452). After several washes with PBS, samples were treated with anti-DMP1 antibodies at 20 pg/ml for 2 hours at room temperature, then washed several times with PBS. A brown chromogenic pigment dependent on the presence of the bound detection reagent/anti-DMP1 antibody/bone-DMP1 complex was developed using the Mouse on Mouse Polymer IHC Kit (Abeam #ab269452). Chromogen development time was determined by comparing the signal intensity to the one of a control adjacent slide incubated without the primary antibody (about two minutes). Chromogen development was halted by rapidly quenching the substrate mixture with distilled water and a final submersion in excess distilled water for 5 minutes at room temperature. The duration of chromogen development was tightly controlled across all slides to reduce slide-to- slide variability. Finally, sections were counterstained using Vector Hematoxylin QS (Vector Laboratories #H-3404) and non-aqueously mounted for brightfield imaging.

Fig. 5 shows the longitudinal interface of bone marrow and the trabecular region (femur ‘cap’) in mouse femur sections stained using an anti-DMP1 monoclonal antibody (mAb), as compared to a control. The anti-DMP1 mAb showed strong immunoreactivity in murine bone tissue. Example 11 : Characterization of ANDbodies binding DMP1 address and SOST target a. In vitro assays for Sclerostin-blocking activity in ANDbodies

The ability of ANDbodies targeting DMP1 and SOST to inhibit SOST activity is determined using TCF/LEF Wnt reporter cells and mineralization of MC3T3 osteoblastic cells as described above (see, e.g., Example 1 (e)). b. Tissue-specific binding of anti-DMP1/SOST ANDbodies

The ability of ANDbodies targeting DMP1 and SOST to bind specifically to bone tissue and inhibit SOST activity is evaluated using IHC to test binding to murine bone, to tissues on a murine tissue microarray (as described above, substituting murine tissues for human tissues), and antihuman IgG 1 secondary antibody for anti-mouse secondary antibody as described above (see, e.g., Example 1 (c)).

Example 12: Design of an ANDbody binding IBSP address and SOST target a. Vaccination to create anti-IBSP antibodies

Antibodies binding to integrin binding sialoprotein (IBSP) are produced as described for CHAD in Example 1 , except that the full-length sequence of human IBSP (protein accession NP_004958.2) is fused to human IgG 1 and used for immunization. The human and equivalent mouse protein sequences are used for identification of IBSP antibodies. b. Selecting for inert anti-IBSP antibodies

Anti-IBSP hybridoma clones produced as described above are further evaluated based on their inability to block mineralization in MC3T3 osteoblastic cells as described above (see, e.g., Example 3(b)).

Anti-IBSP antibodies that do not interfere with mineralization are selected for use in ANDbodies. c. Identifying anti-IBSP antibodies that bind specifically to mouse bone

Binding strength and specificity of anti- IBSP antibodies is evaluated using IHC analysis of human bone tissue and tissue microarrays as described above (see, e.g., Example 1 (c)).

Example 13: Production and characterization of ANDbodies binding IBSP address and SOST target a. Expressing and purifying ANDbodies as bispecific antibodies

ANDbodies comprising bispecific antibodies binding to both IBSP and SOST are expressed and purified as described above (see, e.g., Examples 2A and 8A). b. Simultaneous binding to IBSP and SOST in vitro

The ability of ANDbodies to simultaneously engage with IBSP and SOST is determined and binding affinity for IBSP and SOST is measured using BLI as described above (see, e.g., Example 8B). c. In vitro assays for Sclerostin-blocking activity in ANDbodies

The ability of ANDbodies targeting IBSP and SOST to inhibit SOST activity is determined using TCF/LEF Wnt reporter cells and mineralization of MC3T3 osteoblastic cells as described above (see, e.g., Example 1 (e)). d. Tissue-specific binding of anti-IBSP/SOST ANDbodies

The ability of ANDbodies targeting IBSP and SOST to bind specifically to bone tissue and inhibit SOST activity is evaluated using IHC to test binding to murine bone, to tissues on a murine tissue microarray (as described above, substituting murine tissues for human tissues), and antihuman IgG 1 secondary antibody for anti-mouse secondary antibody as described above (see, e.g., Example 1 (c)).

Example 14: Design of an ANDbody binding TPBG address and SOST target a. Vaccination to create anti-TPBG antibodies

Antibodies binding to trophoblast glycoprotein (TPBG) are produced as described in Example 1 for CHAD, except that the extracellular domain of human TPBG (NCBI protein accession NP 001363851 .1 , positions M1 -H355) is fused to human lgG1 and used for immunization. The human and equivalent mouse protein sequences are used for identification of anti-TPBG antibodies. b. Selecting for inert anti-TPBG antibodies

Anti-TPBG hybridoma clones produced as described above are further evaluated based on their inability to block mineralization in osteoblastic cell lines, as described above (see, e.g., Example 3(b)).

Anti-TPBG antibodies that do not interfere with mineralization or upregulation of osteogenesis markers are selected for use in ANDbodies. c. Identifying anti-TPBG antibodies that bind specifically to mouse bone

Binding strength and specificity of anti-TPBG antibodies is evaluated using IHC analysis of human bone tissue and tissue microarrays, as described above (see, e.g., Example 1 (c)).

Example 15: Evaluation of anti-TPBG antibodies binding to cell surface TPBG

Commercially available antibodies recognizing human and mouse TPBG were tested as candidate ANDbody localizers (address target binding domains) in cell binding assays and using the primary osteoblastic cell line MC3T3.

The anti-TPBG antibody MA5-24228 (Thermo Fisher Scientific) was further characterized for live cell binding using flow cytometry and immunofluorescence of human and mouse transient overexpression in HEK293T cells and the human MC3T3 osteoblast-like cell line, respectively.

Full-length human and mouse TPBG were transiently expressed in HEK 293T cells, followed by flow cytometric quantification of binding of the anti-TPBG antibody to TPBG. Custom expression plasmids encoding public-domain consensus sequences of human and mouse TPBG sequences were obtained from Genscript based on their commercial expression vector that contains a C-terminal intracellular affinity epitope (FLAG-tag) on the same open reading frame as the inserted TPBG sequence. Human or mouse TPBG plasmids were transfected into healthy 293T cells seeded in 96- well plates by complexing with LIPOFECTAMINE™ 3000 according to the manufacturer’s recommendations (Invitrogen #L3000001 ). After 48 hours of presumed TPBG expression, cells were gently harvested using the non-destructive, EDTA-based VERSENE™ cell dissociation reagent (Gibco) and stained for flow cytometry with a defined serial dilution range of a selected anti-TPBG commercial monoclonal or polyclonal antibody, or with an anti-FLAG monoclonal antibody to represent the total amount of recombinant TPBG-FLAG protein present at the time of harvest. After incubation with the primary antibody, cells were stained using the fluorophore-conjugated antibody anti-mouse AF647. For intracellular staining of the FLAG-tag sequence, cells were permeabilized using PERM/WASH™ buffer (BD Bioscience #544723) according to the manufacturer’s protocol and were subsequently stained using the primary antibody rabbit anti-FLAG-AF647. After staining, cells were washed, fixed in 1% PFA, and analyzed by flow cytometry. ZOMBIE AQUA™ was used as viability dye. Median fluorescence intensity of the evaluated antibodies at each antibody concentration in live (non-dead), single cells (double-discriminated) was obtained from FACS analysis using FlowJo software to generate dose-response curves and ECso values.

As shown in Fig. 6, the anti-TPBG monoclonal antibody (mAb) MA5-24228 bound to both murine and human membrane-bound TPBG. Untransfected cell staining and incubation with the secondary antibody alone served as negative controls. Comparable expression levels of FLAG in mouse and human TPBG transfected cells facilitated comparison of antibody binding to mouse and human TPBG.

To assess live cell binding of the anti-TPBG antibody MA5-24228 to endogenous TPBG, immunofluorescence analysis was performed using the osteoblast MC3T3-E1 subclone 4 cell line (ATCC #CRL-2593). 6000 cells were plated in 100 pl complete medium cultured in 96-well glass bottom wells for 9 days. Cells were incubated with anti-TPBG MA5-24228 primary antibody on ice, washed and incubated with secondary antibody goat anti mouse AF647 (A21235) on ice, washed, and lightly fixed in 1% PFA. Qualitative images were obtained using the EVOS™ M7000 imaging system. As shown in Fig. 7, the anti-TPBG antibody MA5-24228 bound to the cell surface of the cells.

Example 16: Production and characterization of ANDbodies binding TPBG address and SOST target a. Expressing and purifying ANDbodies as bispecific antibodies

ANDbodies comprising bispecific antibodies binding to both TPBG and SOST are expressed and purified as described above (see, e.g., Examples 2A and 8A). b. Simultaneous binding to TPBG and SOST in vitro

The ability of ANDbodies to simultaneously engage with TPBG and SOST is determined and binding affinity for TPBG and SOST is measured using BLI as described above (see, e.g., Example 8B). c. In vitro assays for Sclerostin-blocking activity in ANDbodies

The ability of ANDbodies targeting TPBG and SOST to inhibit SOST activity is determined using TCF/LEF Wnt reporter cells and mineralization of MC3T3 osteoblastic cells as described above (see, e.g., Example 1 (e)). d. Tissue-specific binding of anti-TPBG/SOST ANDbodies

The ability of ANDbodies targeting TPBG and SOST to bind specifically to bone tissue and inhibit SOST activity is evaluated using IHC to test binding to murine bone, to tissues on a murine tissue microarray (as described above, substituting murine tissues for human tissues), and antihuman IgG 1 secondary antibody for anti-mouse secondary antibody (see, e.g., Example 1 (c)).

Example 17: Generation of an alendronate-conjugated ANDbody molecule binding the target SOST

To produce antibodies for conjugation to linker-conjugated small molecules, CHO cells were transiently transfected with a 2:3 mass ratio of an S239C FC variant heavy chain: light chain plasmid using PEI and maintained according to the manufacturer’s instructions (37°C, 5% CO2 on a shaking platform). Cultures were fed 4-24 hours post-transfection to a final concentration of 5% v/v Gibco Feed B, 1% v/v L-alanyl-glutamine, and 4 mM valproic acid. After 4-7 days, supernatants were harvested by 0.22 pm filtration. Filtered supernatants were purified via Protein A affinity chromatography (Cytiva 5 mL MABSELECT™ PrismA column) equilibrated in PBS pH 7.4 running buffer and 0.1 M Sodium Citrate pH 3.0 elution buffer. Eluted protein was immediately neutralized with 10% v/v of 1 M sodium acetate, pH 6.0. Protein was further purified via Size Exclusion Chromatography equilibrated in the conjugation formulation buffer of 20 mM His-HAC, 150 mM NaCI, pH 5.5

Antibody constructs binding SOST and DKK1 targets are conjugated to bisphosphonate molecules with different binding potencies to bone mineral hydroxyapatite, including alendronate, risedronate, etidronate, ibandronate, clodronate, tiludronate, pamidronate and zoledronate, using a two-step process, thereby generating ANDbodies.

To generate BIS-L1 (alendronate molecule), first, BCN-PEG4-Maleimide was conjugated an anti-SOST mAb (romosozumab comprising a S239C amino acid substitution mutation) at an S239C site via thiol chemistry. The bisphosphonate molecule was then conjugated to this site via click chemistry, generating the final drug (ANDbody comprising BIS-L1 ) (Fig. 8). a. Characterization of bisphosphonate-conjugated ANDbodies (purity, drug-antibody ratio)

To ascertain the aggregation status of antibody-drug conjugate (ADC) material (M5-D7: an ANDbody comprising an anti-SOST antibody conjugated to alendronate), 20 pL sample was injected onto a 5/150 SUPERDEX™ Increase 200pg column (Cytiva) via HPLC. The column was washed with 1 .2 column volumes formulation buffer to determine elution time of ADC and aggregate.

To ascertain whether the drug-antibody ratio (DAR) of conjugated ANDbody material was within the target range of 2.0 ± 0.2, both hydrophobic interaction chromatography (HIC) and liquid chromatography-mass spectrometry (LC-MS) were employed. For LC-MS, 50 pg of ADCs were prepared at 1 mg/mL concentration with 50 mmol/L DTT and incubated at 37°C to reduce ANDbody material. Samples were injected onto a PLRP-S 1000A, 5um, 2.1 x 50 mm (Agilent) column equilibrated in Mobile Phase Buffer A (100% HPLC grade water, with 0.1% formic acid and 0.025% trifluoroacetic acid) and washed at a 27-49-95% gradient of Mobile Phase buffer B (100% acetonitrile, with 0.1% formic acid and 0.025% trifluoroacetic acid) at 70°C. LC data were deconvoluted to calculate drug-to-antibody ratio (DAR) from each peak.

To ascertain DAR via HIC, samples were diluted into HIC buffer A (25 mM Potassium Phosphate pH 7, 1 .5 M Ammonium Sulphate) and loaded onto the HIC column (HIC column, TSKgel Butyl NPR, 4.8 mm x 3.5 cm). The column was washed with 0-100% HIC buffer B (25 mM Potassium Phosphate pH 7) over 12 minutes at 1 mL/minute flow rate. Eluted peaks were analyzed against parental ANDbody HIC elution to determine DAR.

To ascertain endotoxin levels, a Nexgen multi-cartridge PTS system (Charles River Laboratories) was loaded with 25 pL ANDbody and ADC material in each cartridge well, as per manufacturer instructions.

Results are shown in Table 4. Conjugation of the anti-SOST antibody and alendronate to create an ANDbody was successful: the DAR was within the acceptable range, there was no perceived change in aggregation status, and the endotoxin level was low enough to enable in vivo experiments.

Table 4: ANDbody Characterization Data

b. Kinetic measurement of binding of alendronate-conjugated ANDbody molecule to target protein in BLI

In vitro characterization of the M5-D7 ANDbody and control antibodies by BLI was performed as previously described (Table 5). The parental antibodies PRO136 (humanized anti-SOST mAb) and PRO236 (murine parental version of PRO136 before humanization) were used as are control antibodies. Molecules were assessed for binding to human SOST (hSOST) and mouse SOST (mSOST). Conjugation of alendronate did not disrupt or alter the kinetic parameters of the parental antibodies.

Table 5: Kinetic parameters of an anti-SOST bisphosphonate conjugate

Example 18: A bisphosphonate-conjugated anti-SOST ANDbody blocks Sclerostin in cellbased assay

The bisphosphonate-conjugated anti-SOST ANDbody (M5-D7; also referred to herein as anti- SOST-BIS-L1 (2), wherein “2” indicates a drug-antibody ratio of two alendronate molecules per anti- SOST mAb molecule) was tested for its ability to block sclerostin and restore Wnt1 signaling in vitro in a cell-based assay using a commercial HEK293 reporter cell line (BPS Biosciences #60501 ). Prior to testing, the ECso concentration of Wnt1 (R&D Systems #R&D 9765-WN-010) and IC50 of SOST (Aero Biosystems #HST-H5245) against Wnt1 EC50 were determined for the vendor’s recommended assay protocol. This involved seeding and treating 35,000 cells per 96-well plate with 10mM LiCI, followed Wnt1 stimulation for 6 hours on the next day to induce luciferase expression, followed by endpoint cell lysis and quantification of luminescence (ONE-STEP™ Luciferase Assay System, BPS Biosciences #60690-2). Following assay calibration, a dose-response analysis was performed using increasing amounts of the bisphosphonate-conjugated anti-SOST ANDbody (M5-D7) in the same assay format by adding a pre-incubation step in which the ANDbody was mixed with the IC50 concentration of SOST before mixture with Wnt1 EC50 and 6 hours of stimulation of the reporter cells. Results demonstrated that the bisphosphonate-conjugated anti-SOST ANDbody retained the ability to block SOST with equal potency compared with unconjugated anti-SOST parental mAbs (Fig. 9). Example 19: Rapid in vitro binding of an alendronate-conjugated ANDbody molecule to hydroxyapatite

Bone homing capability of the bisphosphonate-conjugated anti-SOST ANDbody (M5-D7) was assessed in binding assays using hydroxyapatite (Hap) crystals, the mineral component of bone tissue.

Physiologically relevant concentrations (~1 nM) of anti-SOST (PRO136) or anti-SOST-BIS- L1 (2) (M5-D7) were incubated with excess hydroxyapatite powder (Sigma 900204, average particle size, 5 pm) to model binding capacity to bone matrix in vivo. Samples were mixed in a starting total volume of 5 ml in 5 ml protein LoBind tubes (EPPENDORF®) and kept rotating in a dry 37°C incubator. Periodically, the unbound mAb remaining in the solution was collected by separating from the hydroxyapatite crystals with the bound hlgG using a 0.2 micron spin filter and stored at 4°C until all time points were collected. Percentage of mAb binding over time was calculated via sandwich ELISA by measuring free (unbound) human IgG remaining in solution, as compared to a solution in which hlgG was incubated without hydroxyapatite. Exact concentrations were determined using a standard curve using the same PRO136 hlgG antibody (Capture and detection antibodies: Jackson ImmunoResearch #109-005-170, BIO-RAD #STAR127P. TMB Substrate: Thermo Scientific # 34028). Anti-SOST-BIS-L1 (2) was completely depleted from solution within 2 hours of incubation with hydroxyapatite, suggesting high affinity of bisphosphonate for bone mineral, as compared to the unconjugated anti-SOST (PRO136) antibody (Fig. 10).

Example 20. Pharmacokinetics and distribution of bone-targeted ANDbodies

Mice were injected subcutaneously (SQ) with 20 milligrams per kilogram of body weight (mpk) of the bisphosphonate-conjugated anti-SOST ANDbody (M5-D7), an unconjugated anti-SOST antibody, or a control anti-RSV antibody. Femurs were collected at Days 1 , 3 and 8 and were fixed, decalcified and embedded in paraffin blocks. Femur sections were stained using anti-human IgG and developed using VECTOR® Red substrate (VectorLabs). Sections were imaged using EVOS™ M7000i Cell Imaging System and analyzed using the HALO® image analysis platform. Bisphosphonate conjugation resulted in significant accumulation in bone cortex (i.e., enrichment in the bone compartment) that is readily observed on Day 1 and increases over time compared to control antibody molecules (Figs. 11 A and 11 B).

Example 21 : In vivo effect of bone-targeted ANDbodies: P1NP changes in mice after administration of bone-targeted ANDbody constructs

Mice were injected SQ with 20 mpk of the bisphosphonate-conjugated anti-SOST ANDbody (M5-D7), an unconjugated anti-SOST antibody (PRO136), or a control anti-RSV antibody (PRO022). Serum samples were obtained at Days 1 , 3 and 8, and serum levels of the bone turnover marker procollagen type 1 N-terminal propeptide (P1 NP) were evaluated using ELISA (Abeam # ab210579). Increases in levels of P1 NP were observed using the bisphosphonate-conjugated anti-SOST ANDbody at a similar rate to the unconjugated anti-SOST antibody, demonstrating efficacy of the M5- D7 ANDbody construct (Fig. 12).

Example 22: Design of an ANDbody construct binding Dickkopf-1 (DKK1) and SOST as a bispecific actuator conjugated with a bisphosphonate bone localizer targeting to hydroxyapatite address a. Generation of anti-DKK1 and anti-DKK1 anti-SOST bispecific antibody constructs Anti-DKK1 antibody constructs and anti-DKK1 anti-SOST bispecific antibodies are generated utilizing the protocols provided herein. b. Generation of anti-DKK1 and anti-DKK1 anti-SOST bispecific antibody constructs conjugated to bisphosphonate

Anti-DKK1 antibodies and anti-DKK1 anti-SOST bispecific antibody constructs are conjugated to a bisphosphonate molecule as described above (see, e.g., Example 17), thereby generating (i) an ANDbody comprising an anti-DKK1 antibody conjugated to a bisphosphonate molecule; and (ii) an ANDbody comprising an anti-DKK1 anti-SOST bispecific antibody conjugated to a bisphosphonate molecule. c. In vitro characterization of anti-DKK1 anti-SOST bispecific antibody constructs

In vitro characterization of the anti-DKK1 antibody and anti-DKK1 anti-SOST bispecific antibody was performed (Table 6). PRO136 and PRO236 were synthesized as previously stated and are control antibodies. Molecules were assessed for binding to human DKK1 (hDKK1 ) and human SOST (hSOST).

Table 6: Kinetic parameters of anti-DKK1 mAb and anti-DKK1 anti-SOST bispecific antibody

Example 23: Assessment of anti-DKK1 and anti-DKK1 anti-SOST ANDbodies in in vitro cellbased assays

Purified ANDbody constructs, mAbs, and bispecific antibodies are tested for their ability to block DKK1 function in in vitro cell-based assays using HEK293T cells as described above see, e.g., (Example 1 (e)).

VII. OTHER EMBODIMENTS

Some embodiments of the technology described herein can be defined according to any of the following numbered embodiments:

1 . A macromolecule comprising a first binding site and a second binding site, wherein:

(a) the first binding site is specific for an effector target in a bone tissue or bone cell in a subject, and

(b) the second binding site is specific for an address target expressed in a bone tissue or bone cell in the subject; wherein:

(i) the second binding site localizes the first binding site to the address target such that the first binding site influences effector target signaling in the bone tissue or bone cell;

(ii) the second binding site does not substantially influence signaling upon binding the address target; and

(iii) the first binding site does not substantially influence effector target signaling in the absence of localization by the second binding site.

2. A macromolecule comprising a first binding site and a second binding site, wherein:

(iii) the first binding site does not substantially influence effector target signaling in the absence of localization by the second binding site; and wherein localization of the macromolecule to a non-target tissue or cell is substantially reduced relative to localization of a reference macromolecule lacking the second binding site.

3. A macromolecule comprising a first binding site and a second binding site, wherein:

(iii) the first binding site does not substantially influence effector target signaling in the absence of localization by the second binding site; and wherein localization of the macromolecule to the bone tissue or bone cell is substantially increased relative to localization of a reference macromolecule lacking the second binding site.

4. A macromolecule comprising a first binding site and a second binding site, wherein:

(iii) the first binding site does not substantially influence effector target signaling in the absence of localization by the second binding site; and wherein at least 25% of the macromolecule administered to a subject is detected at the bone tissue or bone cell at a time point between 1 and 7 days following administration.

5. A macromolecule comprising a first binding site and a second binding site, wherein:

(iii) the first binding site does not substantially influence effector target signaling in the absence of localization by the second binding site; and wherein the affinity of the first binding site for the effector target is lower than the affinity of the second binding site for the address target.

6. A macromolecule comprising a first binding site and a second binding site, wherein:

(a) the first binding site is specific for an effector target in a bone tissue or bone cell in a subject, and (b) the second binding site is specific for an address target expressed in a bone tissue or bone cell in the subject; wherein:

(iii) the first binding site does not substantially influence effector target signaling in the absence of localization by the second binding site; and wherein the avidity of the first binding site for the effector target is lower than the avidity of the second binding site for the address target.

7. A macromolecule comprising a first binding site and a second binding site, wherein:

(i) the second binding site localizes the first binding site to the address target such that the first binding site influences effector target signaling in the target tissue or cell;

(iii) the first binding site does not substantially influence effector target signaling in the absence of localization by the second binding site; and wherein the potency of the first binding site at the bone tissue or cell is substantially increased relative to a reference macromolecule lacking the second binding site.

8. The macromolecule of any one of embodiments 1 -7, wherein the first binding site has a low affinity for the effector target.

9. The macromolecule of any one of embodiments 1 -7, wherein the first binding site has a low avidity for the effector target.

10. The macromolecule of any one of embodiments 1 -4 and 6-9, wherein the affinity of the first binding site for the effector target is lower than the affinity of the second binding site for the address target.

11 . The macromolecule of any one of embodiments 1 -10, wherein the avidity of the first binding site for the effector target is lower than the avidity of the second binding site for the address target.

12. The macromolecule of any one of embodiments 1 -11 , wherein: (a) the Kd of the first binding site for the effector target is higher than the Kd of the second binding site for the address target;

(b) the EC50 of the first binding site for the effector target is higher than the EC50 of the second binding site for the address target; or

(c) the IC50 of the first binding site for the effector target is higher than the IC50 of the second binding site for the address target.

13. The macromolecule of any one of embodiments 1 -12, wherein the first binding site has an affinity to the effector target of at least about 2 times, at least about 5 times, or at least about 10 times less than the affinity of the second binding site to the address target.

14. The macromolecule of any one of embodiments 1 -13, wherein the affinity of the second binding site to the address target has a Kd of greater than about 1 nM, greater than about 2 nM, or greater than about 50 nm.

15. The macromolecule of any one of embodiments 1 -14, wherein the effector target is a protein, lipid, or sugar.

16. The macromolecule of any one of embodiments 1 -15, wherein the effector target is a cell membrane-associated target.

17. The macromolecule of embodiment 15 or 16, wherein the effector target is a protein.

18. The macromolecule of embodiment 17, wherein the effector target is a secreted protein.

19. The macromolecule of any one of embodiments 1 -18, wherein the macromolecule agonizes the effector target.

20. The macromolecule of any one of embodiments 1 -18, wherein the macromolecule antagonizes the effector target.

21 . The macromolecule of any one of embodiments 1 -20, wherein the address target is a protein, lipid, or sugar.

22. The macromolecule of embodiment 21 , wherein the address target is a protein.

23. The macromolecule of any one of embodiments 17-22, wherein expression of the effector target or the address target is expression of an RNA sequence encoding the effector target or the address target. 24. The macromolecule of embodiment 23, wherein the expression level of the effector target or the address target is assessed by using a RNA sequence dataset.

25. The macromolecule of embodiment 24, wherein the RNA sequence dataset is a Genotype- Tissue Expression (GTEx) dataset or a Human Protein Atlas (HPA) dataset.

26. The macromolecule of embodiment 22, wherein expression of the effector target or the address target is protein expression.

27. The macromolecule of any one of embodiments 1 -26, wherein the effector target is systemically expressed in the subject.

28. The macromolecule of any one of embodiments 1 -26, wherein the effector target is regionally expressed in the subject.

29. The macromolecule of any one of embodiments 1 -26, wherein the effector target is locally expressed in the subject.

30. The macromolecule of any one of embodiments 1 -29, wherein the address target is regionally expressed in the subject.

31 . The macromolecule of any one of embodiments 1 -29, wherein the address target is locally expressed in the subject.

32. The macromolecule of any one of embodiments 1 -29, wherein the expression of the address target is restricted to a cell type in the subject.

33. The macromolecule of any one of embodiments 1 -32, wherein the address target is a soluble protein or an extracellular matrix (ECM)-associated protein and is not present in detectable amounts on the cell surface.

34. The macromolecule of embodiment 33, wherein the address target is expressed in the ECM and is not present in detectable amounts elsewhere in the subject.

35. The macromolecule of any one of embodiments 1 -34, wherein the address target is expressed only by a cell in the subject when in a specific cell state.

36. The macromolecule of any one of embodiments 1 -35, wherein the address target is expressed only by a cell in the subject when in a disease state. 37. The macromolecule of any one of embodiments 1 -36, wherein the address target is not expressed in a tissue in which binding of the second binding site to the effector target is deleterious to the subject.

38. The macromolecule of any one of embodiments 1 -37, wherein the binding site for the address target does not bind in detectable amounts to the binding site of a natural ligand of the address target.

39. The macromolecule of any one of embodiments 1 -38, wherein expression of the address target is substantially higher in bone tissue or bone cells than in any other tissue or cell type.

40. The macromolecule of any one of embodiments 1 -39, wherein the effector target and address target are on the same cell.

41 . The macromolecule of any one of embodiments 1 -39, wherein the effector target and address target are on different cells.

42. The macromolecule of embodiment 41 , wherein the effector target and address target are on different cells of the same cell type.

43. The macromolecule of embodiment 41 , wherein the effector target and address target are on different cells of different cell types.

44. The macromolecule of any one of embodiments 40-43, wherein the effector target and address target are on different cells located within 100 nm of each other in the subject.

45. The macromolecule of any one of embodiments 40-43, wherein either the effector target or the address target is present on a cell surface.

46. The macromolecule of any one of embodiments 1 -45, wherein the macromolecule is a DNA polynucleotide.

47. The macromolecule of any one of embodiments 1 -45, wherein the macromolecule comprises an RNA or RNA-polypeptide conjugate.

48. The macromolecule of any one of embodiments 1 -45 and 47, wherein the macromolecule comprises a polypeptide.

49. The macromolecule of any one of embodiments 1 -45, wherein the macromolecule is a polypeptide. 50. The macromolecule of embodiment 48 or 49, wherein the polypeptide is an antibody or antigen-binding fragment thereof.

51 . The macromolecule of embodiment 50, wherein the first binding site and the second binding site each comprise a VH and/or a VL.

52. The macromolecule of embodiment 51 , wherein the macromolecule is an antibody comprising a first binding site that is specific for the effector target in the subject and a second binding site that is specific for the address target.

53. The macromolecule of embodiment 51 or 52, wherein the macromolecule is an asymmetric antibody or a symmetric antibody.

54. The macromolecule of any one of embodiments 50-53, wherein the antibody or antigenbinding fragment thereof comprises an scFv, BsigG, a BsAb fragment, a BiTE, a dual-affinity retargeting protein (DART), a tandem diabody (TandAb), a diabody, an Fab2, a di-scFv, chemically linked F(ab’)2, an Ig molecule with 2, 3 or 4 different antigen binding sites, a DVI-IgG four-in-one, an ImmTac, an HSAbody, an IgG-IgG, a Cov-X-Body, an scFv1 -PEG-scFv2, an appended IgG, an DVD- IgG, an affibody, an affilin, an affimer, an affitin, an alphabody, an anticalin, an avimer, a DARPin, a Fynomer, a monobody, a nanoCLAMP, a bis-Fab, an Fv, a Fab, a Fab’-SH, a linear antibody, an scFv, an antibody with only a heavy chain (Humabody), an ScFab, an IgG antibody fragment, a single-chain variable region antibody, a single-domain heavy chain antibody, a bispecific triplebody, a BiKE, a CrossMAb, a dsDb, an scDb, tandem a dAb I VHH, a triple dAb VHH, a tetravalent dAb I VHH, a Fab-scFv, a Fab-Fv, or a DART-Fc, an adnectin, a Kunitz-type inhibitor, or a receptor decoy.

55. The macromolecule of embodiment 48, wherein the polypeptide is a ligand of the effector target or a ligand of the address target.

56. The macromolecule of embodiment 55, wherein the ligand is a natural ligand, a modified ligand, or a synthetic ligand.

57. The macromolecule of embodiment 55 or 56, wherein the effector target or address target is a receptor and the polypeptide is a ligand thereof.

58. The macromolecule of any one of embodiments 55-57, wherein the first binding site comprises an antibody or antigen-binding fragment thereof and the second binding site comprises a ligand of the address target. 59. The macromolecule of any one of embodiments 55-57, wherein the first binding site comprises a ligand of the effector target and the second binding site comprises an antibody or antigen-binding fragment thereof.

60. The macromolecule of any one of embodiments 1 -45 and 48-59, wherein the amino acid sequences of the first and second binding sites are at least about 10% identical, at least about 20% identical, at least about 30% identical, at least about 40% identical, at least about 50% identical, at least about 60% identical, or at least about 70% identical.

61 . The macromolecule of any one of embodiments 1 -48, wherein the first binding site comprises an antibody or antigen-binding fragment thereof and the second binding site comprises a small molecule that binds to the address target.

62. The macromolecule of any one of embodiments 1 -48, wherein the first binding site comprises a small molecule that binds to the effector target and the second binding site comprises an antibody or antigen-binding fragment thereof.

63. The macromolecule of any one of embodiments 1 -62, wherein the address target has a Gini coefficient higher than about 0.4, about 0.5, about 0.57, about 0.65, about 0.7, about 0.85, about 0.90, or about 0.95.

64. The macromolecule of any one of embodiments 1 -63, wherein the address target has a Tau coefficient higher than about 0.67, about 0.75, about 0.8, about 0.85, about 0.90, or about 0.95.

65. The macromolecule of any one of embodiments 1 -64, wherein the effector target has a Gini coefficient lower than about 0.25, about 0.20, or about 0.15.

66. The macromolecule of any one of embodiments 1 -65, wherein the effector target has a Tau coefficient lower than about 0.25, about 0.20, or about 0.15.

67. The macromolecule of any one of embodiments 1 -66, further comprising a third binding site.

68. The macromolecule of embodiment 67, wherein the third binding site is the same as the first binding site.

69. The macromolecule of embodiment 67, wherein the third binding site is the same as the second binding site.

70. The macromolecule of any one of embodiments 1 -69, wherein the first binding site and second binding site are directly joined to each other in the macromolecule. 71 . The macromolecule of any one of embodiments 1 -70, wherein the first binding site and the second binding site in the macromolecule are joined by a stable domain.

72. The macromolecule of any one of embodiments 1 -71 , wherein the address target is encoded by a gene selected from the group consisting of the genes recited in Table 2.

73. The macromolecule of any one of embodiments 1 -72, wherein the address target is chondroadherin (CHAD).

74. The macromolecule of embodiment 73, wherein the second binding site is an anti-CHAD antibody or antigen-binding fragment thereof.

75. The macromolecule of embodiment 74, wherein the anti-CHAD antibody or antigen-binding fragment thereof does not substantially block chondrocyte alpha2 betal integrin dependent adhesion.

76. The macromolecule of any one of embodiments 1 -72, wherein the address target is dentin matrix acidic phosphoprotein 1 (DMP1 ).

77. The macromolecule of embodiment 76, wherein the second binding site is an anti-DMP1 antibody or antigen-binding fragment thereof.

78. The macromolecule of embodiment 77, wherein the anti-DMP1 antibody or antigen-binding fragment thereof does not substantially block mineralization by osteoblastic cells.

79. The macromolecule of any one of embodiments 1 -72, wherein the address target is integrin binding sialoprotein (IBSP).

80. The macromolecule of embodiment 79, wherein the second binding site is an anti-IBSP antibody or antigen-binding fragment thereof.

81 . The macromolecule of embodiment 80, wherein the anti-IBSP antibody or antigen-binding fragment thereof does not substantially block mineralization by osteoblastic cells.

82. The macromolecule of any one of embodiments 1 -72, wherein the address target is trophoblast glycoprotein (TPBG).

83. The macromolecule of embodiment 82, wherein the second binding site is an anti-TPBG antibody or antigen-binding fragment thereof.

84. The macromolecule of embodiment 83, wherein the anti-TPBG antibody or antigen-binding fragment thereof does not substantially block mineralization by osteoblastic cells. 85. The macromolecule of any one of embodiments 1 -72, wherein the address target is hydroxyapatite.

86. The macromolecule of embodiment 85, wherein the second binding site is a bisphosphonate, optionally wherein the bisphosphonate is alendronate, risedronate, etidronate, ibandronate, clodronate, tiludronate, pamidronate, or zoledronate.

87. The macromolecule of any one of embodiments 1 -86, wherein the address target is interferon- induced transmembrane protein 5 (IFITM5).

88. The macromolecule of embodiment 87, wherein the second binding site is an anti-IFITM5 antibody or antigen-binding fragment thereof.

89. The macromolecule of embodiment 88, wherein the anti- IFITM5 antibody or antigen-binding fragment thereof does not substantially interfere with expression of one or more osteogenesis-related genes.

90. The macromolecule of embodiment 89, wherein the one or more osteogenesis-related genes include one or more of Runx2, ALP and OCN.

91 . The macromolecule of any one of embodiments 87-90, wherein the anti-IFITM5 antibody or antigen-binding fragment thereof does not substantially block mineralization by osteoblastic cells.

92. The macromolecule of any one of embodiments 1 -91 , wherein the effector target is sclerostin (SOST).

93. The macromolecule of embodiment 92, wherein the first binding site is an anti-SOST antibody or antigen-binding fragment thereof.

94. The macromolecule of embodiment 92, wherein the first binding site comprises a SOST inhibitor.

95. The macromolecule of embodiment 94, wherein the SOST inhibitor is romosozumab, blosozumab, setrusumab, or SHR-1222 or an antigen-binding fragment thereof.

96. The macromolecule of any one of embodiments 1 -86, wherein the effector target is dickkopf-1 (DKK1 ).

97. The macromolecule of embodiment 96, wherein the first binding site is an anti-DKK1 antibody or antigen-binding fragment thereof. 98. The macromolecule of embodiment 96 or 97, wherein the first binding site comprises a DKK1 inhibitor.

99. The macromolecule of any one of embodiments 1 -72, wherein the first binding site comprises a receptor activator of nuclear factor kappa-p ligand (RANKL) inhibitor.

100. The macromolecule of embodiment 99, wherein the RANKL inhibitor is denosumab or an antigen-binding fragment thereof.

101 . The macromolecule of any one of embodiments 1 -72, wherein the first binding site comprises a cathepsin K inhibitor.

102. The macromolecule of embodiment 101 , wherein the cathepsin K inhibitor is odanacatib.

103. The macromolecule of any one of embodiments 1 -102, wherein the subject is a human.

104. A method of delivering a moiety to a bone tissue or bone cell in a subject, comprising administering to the subject a macromolecule of any one of embodiments 1 -103, wherein the bone tissue comprises the address target.

105. The method of embodiment 104, wherein the moiety is a molecule.

106. The method of embodiment 104 or 105, wherein the moiety is not a toxin.

107. The method of embodiment 104, wherein the moiety is a cell.

108. The method of embodiment 107, wherein the moiety is not a T cell or an NK cell.

109. A method of modulating an effector target in a bone tissue, comprising administering to the tissue a macromolecule of any one of embodiments 1 -103, wherein the bone tissue comprises the address target and the effector target.

110. A method of modulating a bone tissue in a subject, comprising administering to the subject a macromolecule of any one of embodiments 1 -103, wherein the bone tissue comprises the address target and the effector target.

111 . A method of treating a subject having a disease or condition associated with an effector target, comprising administering to the subject a macromolecule of any one of embodiments 1 -104, wherein the first binding site of the macromolecule binds the effector target. 112. A macromolecule comprising a first binding site and a second binding site, wherein:

(b) the second binding site is specific for an address target expressed in a bone tissue or bone cell in the subject; wherein the second binding site localizes the first binding site to the address target such that the first binding site influences effector target signaling in the bone tissue or bone cell, wherein the first binding site does not substantially influence effector target signaling in the absence of localization by the second binding site, and wherein the second binding site does not bind to the binding site of the natural ligand of the address target.

113. A macromolecule comprising a first binding site and a second binding site, wherein:

(b) the second binding site is specific for an address target expressed in a bone tissue or bone cell in the subject; wherein the second binding site localizes the first binding site to the address target such that the first binding site influences effector target signaling in the bone tissue or bone cell, wherein the first binding site does not substantially influence effector target signaling in the absence of localization by the second binding site, and wherein the first binding site and second binding site are directly joined to each other in the macromolecule.

114. A macromolecule comprising a first binding site and a second binding site, wherein:

(b) the second binding site is specific for an address target expressed in a bone tissue or bone cell in the subject; wherein the second binding site localizes the first binding site to the address target such that the first binding site influences effector target signaling in the bone tissue or bone cell, wherein the first binding site does not substantially influence effector target signaling in the absence of localization by the second binding site, and wherein the first binding site and second binding are joined to each other by a stable domain.

115. A macromolecule comprising a first binding site and a second binding site, wherein:

(a) the first binding site is specific for an effector target in a bone tissue or bone cell in a subject, and (b) the second binding site is specific for an address target expressed in a bone tissue or bone cell in the subject; wherein the second binding site localizes the first binding site to the address target such that the first binding site influences effector target signaling in the bone tissue or bone cell, wherein the first binding site does not substantially influence effector target signaling in the absence of localization by the second binding site, and wherein the effector target and/or the address target is expressed on a structural tissue in a host.

1 16. A pharmaceutical composition comprising the macromolecule of any one of embodiments 1 - 103 and 1 12-1 15.

1 17. A pharmaceutical composition comprising a macromolecule and one or more pharmaceutically acceptable excipients, wherein the macromolecule comprises a first binding site and a second binding site, wherein:

(b) the second binding site is specific for an address target expressed in a bone tissue or bone cell in the subject; wherein the second binding site localizes the first binding site to the address target such that the first binding site influences effector target signaling in the bone tissue or bone cell, and wherein the first binding site does not substantially influence effector target signaling in the absence of localization by the second binding site.

1 18. The pharmaceutical composition of embodiment 1 16 or 1 17, wherein the pharmaceutical composition is an RNA pharmaceutical composition.

1 19. The pharmaceutical composition of any one of embodiments 1 16-1 18, further comprising a carrier.

120. The pharmaceutical composition of embodiment 1 19, wherein the carrier is a lipid nanoparticle.

121 . The pharmaceutical composition of embodiment 1 19, wherein the carrier is a viral vector.

122. The pharmaceutical composition of embodiment 1 19, wherein the carrier is a membranebased carrier.

123. The pharmaceutical composition of embodiment 1 19, wherein the membrane-based carrier is a cell. 124. The pharmaceutical composition of embodiment 119, wherein the membrane-based carrier is a vesicle.

125. A method for modulating activity of an effector target in the bone of a subject, the method comprising administering to the subject a macromolecule comprising a first binding site and a second binding site, wherein:

(a) the first binding site is specific for an effector target in the subject, and

(b) the second binding site is specific for CHAD.

126. A method for modulating activity of an effector target in the bone of a subject, the method comprising administering to the subject a macromolecule comprising a first binding site and a second binding site, wherein:

(b) the second binding site is specific for IFITM5.

127. A method of localizing a macromolecule at a bone tissue or cell of a subject, the method comprising administering to the subject a macromolecule comprising a first binding site and a second binding site, wherein:

(a) the first binding site is specific for an effector target in a bone tissue or bone cell in the subject, and

(iii) the first binding site does not substantially influence effector target signaling in the absence of localization by the second binding site; and allowing the macromolecule to localize at the bone tissue or bone cell of the subject.

128. A method of concentrating a macromolecule in a bone tissue or cell in a subject, the method comprising administering to the subject a macromolecule comprising a first binding site and a second binding site, wherein:

(iii) the first binding site does not substantially influence effector target signaling in the absence of localization by the second binding site; and allowing the macromolecule to concentrate at the bone tissue or bone cell of the subject, wherein at least 25% of the macromolecule detectable in the subject is detected at the bone tissue or bone cell at a time point between 1 and 7 days following administration of the macromolecule to the subject.

129. The method of embodiment 127 or 128, wherein the potency of the first binding site at the bone tissue or bone cell is substantially increased relative to a reference macromolecule lacking the second binding site.

130. The method of embodiment 127 or 128, wherein effector target signaling by the macromolecule in a non-target tissue or cell of the subject is substantially decreased relative to a reference macromolecule lacking the second binding site.

131 . The method of any one of embodiments 125-130, wherein the macromolecule is a macromolecule of any one of embodiments 1 -103.

132. The macromolecule of any one of embodiments 1 -103 and 112-115, wherein the macromolecule is linked to a small molecule.

133. The macromolecule, method, or pharmaceutical composition of any one of embodiments 1 - 132, wherein the macromolecule and the small molecule are connected by a linker.

134. The macromolecule, method, or pharmaceutical composition of embodiment 133, wherein the linker is a cleavable linker.

135. The macromolecule, method, or pharmaceutical composition of embodiment 133, wherein the linker is a non-cleavable linker.

136. The macromolecule, method, or pharmaceutical composition of any one of embodiments 1 - 135, wherein a single small molecule is linked to the macromolecule.

137. The macromolecule, method, of pharmaceutical composition of any one of embodiments 1 - 135, wherein multiple small molecules are linked to the macromolecule.

138. The macromolecule, method, or pharmaceutical composition of embodiment 137, wherein each of the small molecules is the same. 139. The macromolecule, method, or pharmaceutical composition of embodiment 137, wherein at least two of the small molecules are different from each other.

140. The macromolecule, method, or pharmaceutical composition of any one of embodiments 132-139, wherein the small molecule is odanacatib.

141 . The macromolecule, method, or pharmaceutical composition of any one of embodiments 132-139, wherein the small molecule is a bisphosphonate.

142. The macromolecule, method, or pharmaceutical composition of embodiment 141 , wherein the bisphosphonate is alendronate, risedronate, etidronate, ibandronate, clodronate, tiludronate, pamidronate, or zoledronate.

143. A macromolecule comprising a first binding site and a second binding site, wherein:

(a) the first binding site is specific for sclerostin (SOST), and

(b) the second binding site is specific for chondroadherin (CHAD).

144. A macromolecule comprising a first binding site and a second binding site, wherein:

(a) the first binding site is specific for sclerostin (SOST), and

(b) the second binding site is specific for interferon-induced transmembrane protein 5

(IFITM5).

145. A macromolecule comprising a first binding site and a second binding site, wherein:

(a) the first binding site is specific for Dickkopf-1 (DKK1 ), and

(b) the second binding site is specific for chondroadherin (CHAD).

146. A macromolecule comprising a first binding site and a second binding site, wherein:

(a) the first binding site is specific for sclerostin (SOST), and

(b) the second binding site is specific for dentin matrix acidic phosphoprotein 1 (DMP1 ).

147. A macromolecule comprising a first binding site and a second binding site, wherein:

(a) the first binding site is specific for sclerostin (SOST), and

(b) the second binding site is specific for integrin binding sialoprotein (IBSP).

148. A macromolecule comprising a first binding site and a second binding site, wherein:

(a) the first binding site is specific for sclerostin (SOST), and

(b) the second binding site is specific for trophoblast glycoprotein (TPBG).

149. A macromolecule comprising a first binding site and a second binding site, wherein: (a) the first binding site is specific for sclerostin (SOST), and

(b) the second binding site is specific for hydroxyapatite.

Although the foregoing invention has been described in some detail by way of illustration and example for purposes of clarity of understanding, the descriptions and examples should not be construed as limiting the scope of the invention.

Claims

WHAT IS CLAIMED IS:

1 . A method of localizing a macromolecule at a bone tissue or bone cell of a subject, the method comprising administering to the subject a macromolecule comprising a first binding site and a second binding site, wherein:

2. The method of claim 1 , wherein at least 25% of the macromolecule detectable in the subject is detected at the bone tissue or bone cell at a time point between 1 and 7 days following administration of the macromolecule to the subject.

3. The method of claim 1 , wherein the potency of the first binding site at the bone tissue or bone cell is substantially increased relative to a reference macromolecule lacking the second binding site.

4. The method of claim 3, wherein the first binding site has a low affinity for the effector target.

5. The method of claim 3, wherein the first binding site has a low avidity for the effector target.

6. The method of claim 1 , wherein the affinity of the first binding site for the effector target is lower than the affinity of the second binding site for the address target.

7. The method of claim 1 , wherein the avidity of the first binding site for the effector target is lower than the avidity of the second binding site for the address target.

8. The method of claim 1 , wherein effector target signaling by the macromolecule in a non-target tissue or cell of the subject is substantially decreased relative to a reference macromolecule lacking the second binding site.

9. The method of claim 1 , wherein the address target is regionally expressed in the subject.

10. The method of claim 1 , wherein the address target is locally expressed in the subject.

11 . The method of claim 1 , wherein the expression of the address target is restricted to a cell type in the subject.

12. The method of claim 1 , wherein the address target is expressed only by a cell in the subject when in a specific cell state.

13. The method of claim 1 , wherein the address target is expressed only by a cell in the subject in a disease state.

14. The method of claim 1 , wherein the first binding site or the second binding site comprises a polypeptide.

15. The method of claim 14, wherein the polypeptide is an antibody or antigen-binding fragment thereof.

16. The method of claim 15, wherein the macromolecule is an antibody comprising a first binding site that is specific for the effector target in the subject and a second binding site that is specific for the address target.

17. The method of claim 14, wherein the polypeptide is a ligand of the effector target or a ligand of the address target.

18. The method of claim 17, wherein:

(a) the first binding site comprises an antibody or antigen-binding fragment thereof and the second binding site comprises a ligand of the address target; or

(b) the first binding site comprises a ligand of the effector target and the second binding site comprises an antibody or antigen-binding fragment thereof.

19. The method of claim 15, wherein:

(a) the first binding site comprises an antibody or antigen-binding fragment thereof and the second binding site comprises a small molecule that binds to the address target; or

(b) the first binding site comprises a small molecule that binds to the effector target and the second binding site comprises an antibody or antigen-binding fragment thereof.

20. The method of claim 1 , wherein the second binding site is specific for chondroadherin (CHAD).

21 . The method of claim 1 , wherein the second binding site is specific for dentin matrix acidic phosphoprotein 1 (DMP1 ).

22. The method of claim 1 , wherein the second binding site is specific for integrin binding sialoprotein (IBSP).

23. The method of claim 1 , wherein the second binding site is specific for trophoblast glycoprotein (TPBG).

24. The method of claim 1 , wherein the second binding site is specific for hydroxyapatite.

25. The method of claim 24, wherein the second binding site is a bisphosphonate.

26. The method of claim 25, wherein the bisphosphonate is alendronate, risedronate, etidronate, ibandronate, clodronate, tiludronate, pamidronate, or zoledronate.

27. The method of any one of claims 20-26, wherein the first binding site is specific for sclerostin (SOST).

28. The method of any one of claims 20-26, wherein the first binding site is specific for dickkopf-1 (DKK1 ).

29. The method of claim 1 , wherein the second binding site is specific for interferon-induced transmembrane protein 5 (IFITM5).

30. The method of claim 29, wherein the first binding site is specific for SOST.

31 . The method of any one of claims 1 -30, wherein the macromolecule is linked to a small molecule.

32. The method of claim 31 , wherein the small molecule is odanacatib.

33. The method of claim 31 , wherein the small molecule is a bisphosphonate.

34. The method of claim 33, wherein the bisphosphonate is alendronate, risedronate, etidronate, ibandronate, clodronate, tiludronate, pamidronate, or zoledronate.

35. The method of any one of claims 30-34, wherein the macromolecule and the small molecule are connected by a linker.

36. The method of claim 35, wherein the linker is a cleavable linker.

37. The method of claim 35, wherein the linker is a non-cleavable linker.