A METHOD OF INTRACELLULAR DELIVERY

Abstract

The present disclosure relates to a method of intracellular delivery of a molecule employing a GLA domain to facilitate entry into the cell.

Claims

1. A method of intra-cellular delivery of a payload said method comprising the step of contacting cells with a molecule comprising: a. a payload linked to a gamma-carboxyglutamic acid component (GLA-component) wherein said GLA-component comprises a GLA domain or an active fragment thereof, and does not comprise an active catalytic domain from a GLA protein.

2. A method according to claim 1, wherein GLA domain or active fragment thereof is independently selected from thrombin, factor VII, factor IX, factor X, protein C, protein S, protein Z, Osteocalcin, Matrix GLA protein, GAS6, Transthretin, Periostin, Proline rich GLA 1, Proline rich GLA 2, Proline rich GLA 3 and Proline rich GLA 4.

3. A method according to claim 1, wherein GLA domain or active fragment thereof is from protein S.

4. A method according to claim 1, wherein the construct comprises an EGF domain selected from thrombin, factor VII, factor IX, factor X, protein C, protein S, protein Z, Osteocalcin, Matrix GLA protein, GAS6, Transthretin, Periostin, Proline rich GLA 1, Proline rich GLA 2, Proline rich GLA 3 and Proline rich GLA 4.

5. A method according to claim 4, wherein the construct comprises an EGF domain selected from protein S.

6. A method according to claim 1, wherein the GLA-component comprises a sequence shown in SEQ ID NO: 6 or a derivative thereof without the his-tag.

7. A method according to claim 1, wherein the GLA-domain component further comprises a Kringle domain.

8. A method according to claim 7, wherein the Kringle domain is from a protein selected from the group comprising Activating transcription factor 2 (ATF); Factor XII (F12); thrombin (F2); Hyaluronan-binding protein 2 (HABP2); Hepatocyte growth factor (HGF); Hepatocyte growth factor activator (HGFAC); Kremen protein 1 (KREMEN1); KREMEN2; Lipoprotein(a) (LPA); LPAL2; Macrophage-stimulating protein (MSP or MST1); Phosphoinositide-3-kinase-interacting protein 1 (PIK3IP1); Tissue plasminogen activator (PLAT); Urokinase (PLAU); Plasmin (PLG); PRSS12; Tyrosine-protein kinase transmembrane receptor ROR1 (ROR1); and Tyrosine-protein kinase transmembrane receptor ROR2 (ROR2).

9. A method according to claim 1, wherein the method is performed in vitro.

10. A method according to claim 1, wherein the intracellular delivery is to a cell in vivo, for example wherein the molecule comprising the GLA component and the payload are administered to a patient.

11. A method according to claim 1, wherein the cell is non-apoptotic.

12. A method according to claim 1, wherein the cell is apoptotic.

13. A method according to claim 1, wherein the cell is a stem cell.

14-17. (canceled)

18. A method according to claim 1, wherein the GLA-component is conjugated to the payload.

19. A method according to claim 1, wherein the payload comprises a drug or biological therapeutic, for example, an anti-viral drug, anti-bacterial drug, anti-parasitic agent, anti-cancer drug, an anti-cancer therapy, such as a chemotherapeutic agent.

20. A method according to claim 1, wherein the payload comprises a toxin, a polymer(for example synthetic or naturally occurring polymers), biologically active proteins (for example enzymes, other antibody or antibody fragments), a drug (small molecule (chemical entity), chemotherapeutic agent), radionuclides (particularly radioiodide, radioisotopes, such as .sup.99mTc) a metal chelating agent, nanoparticles and reporter groups (such as fluorescent or luminescent labels or compounds which may be detected by NMR or ESR spectroscopy).

21. A method according to claim 1, which comprises administering the molecule comprising the GLA component and payload to a cancer patient.

22. A method according to claim 21, wherein the cancer is an epithelial cancer, for example colorectal cancer, testicular cancer, liver cancer, biliary tract cancer, prostate cancer, pancreatic cancer, glioblastoma, melanoma, breast cancer, ovarian cancer, cervical cancer, uterine cancer, gastric cancer, oesophageal cancer, thyroid cancer, renal cancer, bladder cancer, brain cancer, head and neck cancer or lung cancer or alternatively the cancer may be a haematological cancer, for example leukaemia, lymphoma, myeloma and chronic myeloproliferative diseases, such as AML.

23. A method according to claim 19, wherein the drug is an antiviral agent, an anti-parasitic agent or an antibacterial agent.

24-25. (canceled)

26. A method according to claim 1, wherein the payload is converted to an active form inside the cell.

Description

EXAMPLES

[0241] FIG. 1A-D Shows various representations of GLA protein structures.

[0242] FIG. 1E Shows an embodiment of a GLA-component according to the present disclosure.

[0243] FIG. 2 Shows Protein S (PrS) and annexin staining of breast cancer cell lines treated with peroxide to induce apoptosis. A, human MDA-231 cells treated with peroxide and stained with FITC-PrS. B, untreated MDA-231 cells stained as in A. C, treated MDA-231 cells stained with annexin. D, human MCF-7 cells treated with peroxide and stained with PrS. E, murine MET-1 cells, as in D. F, murine 4T1 cells, as in D.

[0244] FIG. 3 Shows overlapping, yet distinct, cellular localization of PrS and annexin. A, murine 4T1 cells treated with peroxide and stained with Cy5 PrS (RED) and FITC annexin (GREEN). Light arrow, co-localized signals; red arrows, cells staining with PrS and not annexin; green arrow, cell staining relatively brighter with annexin but less bright with PrS, indicating distinct binding patterns (insets show PrS and annexin staining separately). B, treated 4T1 cells stained with FITC PrS and Cy5 annexin. Green arrows, cells staining with PrS and not annexin. C, Cy5 annexin staining of treated 4T1 cells pre-incubated with 1,000-fold excess of cold annexin.

[0245] FIG. 4 Shows staining of apoptotic COS-1 cells with PrS and annexin. Cells were treated with t-BHP as described and stained with FITC annexin (left) and Cy5 PrS (right). Arrows indicate subcellular structures presumed to be apoptotic bodies.

[0246] FIG. 5 Shows differential staining of extracellular vesicles with PrS and annexin. Extracellular vesicles were prepared from 4T1 cells and stained with FITC PrS (GREED) and Cy5 annexin (RED). Arrows indicate vesicles staining with annexin only (RED arrow), PrS only (GREEN arrow) and both proteins (light arrow).

[0247] FIG. 6 Shows subcellular localization of PrS and annexin. A, B, apoptotic 4T1 cells were stained with FITC PrS (GREEN arrows) and Cy5 annexin (RED arrows); light arrows, co-localization. C, Possible apoptotic bodies.

[0248] FIG. 7 Shows internalization of PrS within 5 minutes. Apoptotic 4T1 cells were stained with FITC PrS (green) and Cy5 annexin (red) and imaged within 10 min of the addition of the proteins. A, Merged image. B, Hoescht nuclear stain alone.

[0249] FIG. 8 Shows BLI images of 4T1 tumors in mice.

[0250] FIG. 9 SPECT imaging of effect of doxorubicin on 4T1 tumors, using radiolabeled PrS and annexin. Mice with 4T1 breast cancer tumors were imaged with 99mTc PrS (A and B), or annexin (C and D), before (A and C) and 24 h after doxorubicin (B and D).

[0251] FIG. 10 Shows SPECT imaging of cyclohexamide-treated mice. Five mice per panel are shown before (A and C) and 24 h after (B and D) treatment. The mice were imaged with either .sup.99mTc PrS (A and B), or annexin (C and D), Arrows indicate increased liver signal.

[0252] FIG. 11 Shows localization of Cy5 PrS to infected spleen. CD1 mice were infected with bioluminescent Listeria and imaged on day 2 post infection. The mice were injected with Cy5 PrS 30 min before sacrifice, and the spleens removed and frozen. Modestly infected (A) and control uninfected (C) mice are shown. Sections of the infected (B) and uninfected (D) spleens of each mouse in the Cy5 channel are shown, merged with phase contrast images.

[0253] FIG. 12 Shows localization of Cy5 PrS to tumors treated with doxorubicin. Mice implanted with 4T1 breast cancer tumors were treated with doxorubicin (right panels) or left untreated (left panels). 24 hours later the mice were injected intravenously with Cy5 PrS and sacrificed 30 min later. The tumors were removed, frozen, and sectioned for fluorescence microscopy. Merged Cy5/phase contrast images from four different mice are shown.

[0254] FIG. 13 Shows differentiation of TSCs. TSCs were cultured in the presence (left) or absence (right) of growth factors. Arrows in the right panel indicate giant cells characteristic of differentiation.

[0255] FIG. 14 Shows PrS staining of trophoblast stem cells and differentiated trophoblasts. Trophoblast stem cells (left) were differentiated into trophoblast giant cells (right) by withdrawal of growth factors. The cells were stained with Cy5 PrS and imaged.

[0256] FIG. 15 Shows MSC differentiation. MSC were treated as described in the text, for differentiation into adipocytes (upper panels) or osteoblasts (lower panels). Differentiated cells exhibited the expected morphology in each case.

[0257] FIG. 16 Shows MSCs stained with PrS (green), annexin (red), and Hoechst (blue). Cells were imaged within 10 min of addition of the stain mixture.

[0258] FIG. 17 Shows TSCs stained with PrS (green, lightest area), annexin (red, light around the cell membrane), and Hoechst (blue). Cells were imaged within 5 min of addition of the stain mixture.

[0259] FIG. 18 Shows differential staining of TSC vesicles. TSCs were stained as in FIG. 17. The group of cells are secreting large vesicles that stain with annexin (red) and not PrS (green).

[0260] FIG. 19 Shows PrS staining of C17.2 neural progenitor cells. The cells were stained with PrS-FITC and imaged with standard (non-confocal) microscopy.

[0261] FIG. 20 Shows internalization of PrS into TSC at 4C. FITC PrS (green) and Cy5 annexin (red) were added to TSC at 4C and imaged with confocal microscopy.

[0262] FIG. 21 Shows lineage-negative, SCA-1/c-kit staining cells from mouse bone marrow. The cells were not stained with either PI (propidium iodide; to detect dead cells) or PrS at this point in the analysis. Absence of staining for hematopoietic lineages (left panel) and staining of c-kit and SCA1 (right panel) defines the population of HSC, shown in green (lightest areas).

[0263] FIG. 22 PrS staining of long-term HSC. HSC were isolated as in FIG. 1, and stained with FITC PrS. SLAM pattern was determined with Cy7 (x-axis).

[0264] FIG. 23 PrS staining of short-term HSC. HSC were isolated as in FIG. 1, and stained with FITC PrS. SLAM pattern was determined with Cy7 (x-axis).

[0265] FIG. 24 Shows internalization of PrS in long-term HSC. HSC were prepared as described, stained for PrS, and examined with confocal microscopy.

[0266] Green (lightest areas), FITC PrS; blue, Hoescht nuclear stain; red, PI. Note that PI stain is excluded from the nucleus, indicating the cells are alive.

[0267] FIG. 25 Shows an example of dead HSC exhibiting nuclear PI.

[0268] FIG. 26 GLA-mediated uptake is non-toxic to cells.

[0269] This specification also includes sequences 1 to 6, in the associated

SEQUENCE LISTING

[0270] This project initiated the testing of labeled recombinant PrS as an in vivo imaging agent for SPECT (Single Photon Computed Tomography). Surprisingly it was found that the molecule rapidly internalized into apoptotic cells. This unexpected finding led us to explore the phenomenon further, whereupon we found that PrS was also internalized into a subset of non-apoptotic stem cells of several types.

[0271] PrS is protein S GLA domain and protein S EGF domain as shown in SEQ ID NO: 6.

Methods

[0272] For fluorescence, conjugation of Cy5 and FITC was achieved using Amersham (GE Heathcare) and Molecular Probes (Invitrogen) labeling kits, respectively, according to the instructions of the manufacturers. Both kits provide columns for the removal of unconjugated fluorophore. Initially, 0.77 mg of PrS (Fraction 2) in 1 ml and 0.77 mg of annexin in 1 ml were labeled with FITC to test for specificity of binding to apoptotic cells. For co-localization and competition studies 0.68 mg of PrS (Fraction 3) in 1 ml and 0.68 mg of annexin were labeled with Cy5. For confocal microscopy, 0.76 mg of PrS from the second shipment was labeled with FITC and the previously labeled Cy5-conjugated annexin was used. It should be noted that the precise efficiency of labeling was not determined and the recovery from the columns was assumed to be 85%, according to the instructions of the manufacturers of the labeling kits. Thus, the relative staining intensity of the two proteins in any case may reflect these contingencies. The cells were stained for 30 min initially, but it was subsequently determined that less than 5 min was sufficient. To test PrS for apoptotic cell-specificity, four breast cancer cell lines were initially employed; human MDA-231 and MCF7 and murine 4T1 and MET-1. Subsequently, COS-1 monkey kidney cells were also used. Apoptosis was induced with hydrogen peroxide or tertiary-Butyl hydroperoxide (t-BHP). The cells were plated in 24-well plates at 610.sup.4 cells per well or Eppendorf chamber slides at 110.sup.4 cells per well, and apoptosis was induced the next day, using 2 mM H.sub.2O.sub.2, or t-BHP for time points from 30 min to 2 hrs. After induction, the wells were washed with Annexin Binding Buffer (AB; Santa Cruz Biotech), and stained with labeled protein. From past experience and the literature, 5.5 g/ml of annexin protein was used for staining. This amount was adjusted for equimolar addition of PrS by assuming the molecular weights of annexin to be 36 kD and the recombinant PrS to be 30 kD, based on the gel images provided. The cells were stained for 15 min. Hoechst 33342 dye was used for visualizing nucleic acid. The wells were then washed with AB and observed using the EVOS fluorescence microscope while still viable. For confocal microscopy, the Leica SP8 microscope in the Stanford Cell Sciences Imaging Facility was employed. The wells were then washed with AB and observed using the Leica sp8 microscope. Hoechst 33342 dye was used for visualizing nuclei. For toxicity studies, PrS was added to trophoblast stem cells (TSCs) and the viability tested with trypan blue using a Nexcelom Cellometer.

[0273] To test the labeled proteins for the ability to detect tumors, 510.sup.4 4 T1-luc cells were implanted into groups of 5 male BALB/c mice, in the left axillary fat pad. The mice were imaged with in vivo bioluminescence imaging (BLI) each day to monitor tumor growth, starting at 1 week post implantation. The mice were then treated on day 11 post implantation with 13 mg/kg body weight of intraperitoneal (IP) doxorubicin, and BLI was performed the next day. Control mice bearing tumors were left untreated with doxorubicin. 48 hrs post treatment the mice were imaged 1 hr after intravenous tracer injection (anesthesia 1.3 g/kg of urethane IP), with single head A-SPECT gamma camera (Gamma Medica); 1 mm pin hole collimator, 128 steps into a 128128 imaging matrix, 15 seconds per step, 2.7 cm ROR; FOV=upper chest/neck. The injected dose of each protein was 160 l (800 CO. The animals were then sacrificed and biodistribution was performed. For the cyclohexamide treatment experiment, groups of 5 young (7 week old) male Swiss Webster mice were anesthetized (1.3 g/kg of urethane IP) and injected intravenously with 50 mg/kg cycloheximide. 1 hr 45 min after cycloheximide injection, tracer was injected (PrS=180 ul/1.2 mCi per dose; annexin V=170 l/1.05 mCi per dose). 45 min after tracer injection, the mice were imaged with 10 min static whole body images using a single head parallel hole collimator (128128 matrix) on the A-SPECT gamma camera.

[0274] To test for the specific localization of fluorescent PrS to apoptotic sites due to infection in live animals, CD1 mice were injected intravenously with bioluminescent Listeria monocytogenes. This bacterial pathogen infects many organs including the spleen, in which extensive apoptosis of monocytes and granulocytes occurs. At certain times post infection, spleen is the primary site of bacterial replication and so splenic BLI signals from the bacteria can be correlated with the localization of probes for apoptosis. Mice were infected and imaged each day. When splenic signals were evident (day 2 post infection for 210.sup.5 colony forming units of bacteria in 8 week old CD1 female mice), 300 mg/kg body mass of Cy5 PrS was injected into mice, the animals were sacrificed 30 min later, and the spleens removed, frozen in OCT, and sectioned for fluorescence microscopy. Uninfected control mice were employed.

[0275] Flow cytometry was performed by Charles Chan of the Irving Weissman laboratory. Freshly labeled FITC PrS, prepared as described above, was employed. Murine hematopoietic stem cells (HSCs) are routinely purified in this laboratory. The cells were isolated from normal mouse bone marrow by staining for c-Kit+, lineage-negative cells. To further characterize the cells, SLAM marker staining was also performed. These markers stain cells that self-renew and differentiate, whereas non-staining HSCs can only differentiate. Subsequent staining with FITC PrS revealed the percent positive in SLAM-staining cells, as shown in the Results. The cells were then sorted for FITC and examined with confocal microscopy, using Hoechst 33342 for nuclear visualization.

Results

[0276] To assess PrS binding specificity in the context of apoptosis in cell culture, we employed several human and murine breast cancer cell lines. Apoptosis was induced with peroxide as described above, and FITC PrS binding was assessed. Examples of these experiments are shown in FIG. 2. Untreated cells exhibited minimal binding, such as shown in panel B of FIG. 2. Concentrations of peroxide and incubation times were chosen such that only a minority of cells would be affected, because at higher concentrations and/or longer incubation times the cells detached and staining and microscopy was not possible. In addition, the presence of many unaffected cells served as an internal negative control within each field. FITC-annexin showed specificity for apoptosis similar to PrS, serving as an internal positive control. We then tested the two proteins for co-localization and competitive binding. For co-localization, both FITC and Cy5 labeled PrS and annexin were prepared. 4T1 cells were treated with peroxide and stained with Cy5 and FITC labeled PrS and annexin, using both combinations of fluorophores. The cells were then visualized in the EVOS fluorescence microscope. The results are shown in FIG. 3. Under the conditions tested, all the brightly staining cells exhibited staining with both proteins. However, whether using Cy5 or FITC, PrS appeared to stain some cells that annexin did not, albeit weakly (FIG. 3). The relative staining intensity of different cells by each protein sometimes differed between the two probes, i.e., sometimes annexin stained two cells with equal intensity and PrS did not, and vice versa (FIG. 3A, green arrow and insert). Thus, while both probes generally stained the same cells, they appeared to exhibit subtle differences. In the competition assay, increasing excess amounts of unlabeled annexin were pre-incubated with apoptotic 4T1 cells for 15 min and the cells were then stained with Cy5 PrS. Surprisingly, the staining of PrS was not blocked by even 1,000 fold excess of annexin, the highest excess amount tested (FIG. 3C), although these proteins are thought to bind to the same target molecule, exposed PS. Co-staining of annexin and PrS was observed with many cell types. While the two proteins generally stained the same cells in each cell type, other differences became apparent. In particular, some objects smaller than cells were differentially stained (FIG. 4). These objects, which were present in increased numbers after peroxide treatment, were interpreted as apoptotic bodies; membrane-bound cell fragments produced during the fragmentation of apoptotic cells. As shown in FIG. 4, PrS stained these entities, whereas annexin did not, although some of these objects did stain with both proteins. This observation was unexpected. To further explore the differential staining of sub cellular entities, extracellular vesicles (EVs) were prepared from 4T1 murine tumor cells using a standard centrifugation protocol. The two proteins also differentially stained these vesicles (FIG. 5), a result that may have biological and therapeutic implications.

[0277] EVs, specifically exosomes, microvesicles (MVs) and apoptotic bodies (ABs), are presumed to play key roles in cell-cell communication via transfer of biomolecules between cells. The biogenesis of these types of EVs differs, and they originate from either the endosomal (exosomes) or plasma membranes (MV) or are products of programmed cell death (ABs). All mammalian cells are thought to secrete EVs. Each type of EV can transfer molecular cargo to both neighboring and distant cells, affecting cellular behaviors such as those involved in tumor development and progression. In fact, EVs may play a role in nearly all the hallmarks of cancer, including sustaining proliferative signaling, evading growth suppression, resisting cell death, reprogramming energy metabolism, acquiring genomic instability, and developing the tumor microenvironment. They have also been implicated in the induction of angiogenesis, control of invasion, initiation of premetastatic niches, sustaining inflammation, and evading immune surveillance. Immune cells appear to also communicate through EVs and my recognize EVs as signals from tumor cells, infected tissues and wounds. A deeper understanding of the biology of EVs and their contribution to the hallmarks of cancer is leading to new possibilities for diagnosis and treatment of cancer. Development of additional EV surface markers is essential to advancing this field and PrS may be such a determinant.

[0278] Following these studies with fluorescence microscopy, the subcellular localization of the staining by PrS and annexin was then evaluated via confocal microscopy. Murine 4T1 cells (lacking the Luc-GFP reporters) were plated on 8-part chamber slides at 110.sup.4 cells per chamber and apoptosis was induced with 2 mM H.sub.2O.sub.2 or t-BHP (2 hr exposure) the next day. The cells were then washed and stained for 15 min with PrS and annexin. Hoechst 33342 dye was used to stain nucleic acid. In all cases, the most brightly staining cells were stained with both probes. However, in many cells labeled PrS was observed in the cytoplasm, whereas the labeled annexin was not (FIG. 6). Although annexin was internalized and appears in vesicles of a few cells, internalized annexin together with surface localized PrS in the same cell was not observed. These results were unexpected, because the two proteins are both presumed to bind PS. To further study the internalization of PrS, a time course experiment was performed. Apoptotic 4T1 cells were stained for 5 min with Cy5 annexin and FITC PrS, and observed within 5 min of the addition of the probes. PrS was observed in the cytoplasm of these cells immediately, indicating internalization within 5 min (FIG. 7). The time course images also showed that PrS and annexin did not always stain the same cells equally at early time points. The cells in FIG. 7 appear to be in different stages of apoptosis, as the cell on the left shows an uncondensed nucleus surrounded by an apparently intact nuclear membrane, whereas the right cell exhibits the strong staining often characteristic of chromatin condensation that occurs later in the apoptotic process. Staining patterns such as these may indicate that PrS binds earlier in apoptosis than annexin. Although purely conjecture at this point, such a preference would explain many of the differences between these proteins that have been observed so far. For example, the staining of some cells by PrS and not annexin, such as in FIGS. 3A and B may be due to PrS binding earlier in the process of apoptosis. To examine PrS localization in live animals, several experiments were performed. These studies employed chemical and infectious induction of apoptosis in vivo, as well as the localization of PrS to tumors treated with doxorubicin, which is known to induce apoptosis. SPECT imaging using HYNIC-labeled PrS and annexin was performed in animals given 4T1luc breast tumors and treated with doxorubicin. Because the 4T1 tumors have been labeled with luciferase, they can be imaged in mice using in vivo bioluminescence imaging (BLI). One of the images from this experiment is shown in FIG. 8. This method can be used to evaluate tumor implantation and to follow progression in individual animals over time. .sup.99mTc labeled PrS and annexin were then employed for SPECT imaging of animals treated with doxorubicin and controls. An example of the results is shown in FIG. 9. The images of the head and thorax of the two animals show non-specific accumulation of the PrS probe in the salivary gland, and a low signal to noise ratio using this probe. Therefore the threshold of the display in the PrS images shown was lowered to reveal more background, resulting in the brighter false-color of the images. The low signal-to-noise ratio is likely due to HYNIC labeling of only 1 mg of protein, which is sub-optimal, and also due to the inability to perform controlled studies of HYNIC:protein labeling ratio.

[0279] SPECT imaging of mice treated with cyclohexamide, which induces apoptosis in the liver, was also performed (FIG. 10). In FIG. 10, the whole-body images of 5 mice are shown in each panel. As with many radiolabeled probes, background is seen in the kidneys. Treatment of the mice with cyclohexamide increased the annexin SPECT signal in the liver. Again, the PrS showed low signal compared to annexin. Annexin was able to detect the apoptotic livers of cyclohexamide treated mice, whereas PrS showed only slight increase of signal in the liver due to treatment. To test the localization of PrS to apoptotic tissues and treated tumors independently of SPECT imaging and the concomitant complications of HYNIC labeling, mice infected with bacteria that induce apoptotic responses and tumor bearing mice were injected with Cy5 PrS. For infection, we employed Listeria monocytogenes, a bacterial pathogen labeled with luciferase and well characterized for BLI. Characteristic BLI signals from the spleen provide for excellent co-localization studies. CD1 mice were infected as described above and were imaged with BLI on day 2 post infection. The mice were then injected with Cy5 PrS and 30 min later sacrificed, and the spleens removed for sectioning and fluorescence microscopy (FIG. 11). In all cases, splenic sections from infected mice showed much greater Cy5 fluorescence signals than controls. In FIG. 11, the infected mouse shown displayed low photon counts, indicating the infection had not yet progressed very far in this animal. Many mice exhibit 10 times this signal intensity from the spleen on this day. However, the Cy5 channel fluorescence was still very strong relative to the uninfected control shown. This result may reflect the ongoing innate immune response to infection, as granulocytes and macrophages have been shown to be the main source of annexin signal in such animals (these cells are programmed for apoptosis to limit tissue destruction).

[0280] The localization of fluorescent PrS to 4T1 tumors treated with doxorubicin was then tested. Mice implanted with tumors were treated with doxorubicin as described above and Cy5 PrS was injected intravenously 30 min prior to sacrifice and removal of the tumors for sectioning and fluorescence microscopy. The results are shown in FIG. 12. Areas of intense staining were observed in the treated animals, whereas more modest signal was observed from the untreated tumor sections. Although some untreated tumors did exhibit small areas of higher signal than background, no signals of similar intensity to the treated tumors were observed in any of the untreated sections.

[0281] Stem cells are distinct in phenotype from differentiated cells and may express PS non-apoptotically to avoid the induction of immune responses. Trophoblast stem cells (TSCs) differentiate into several types of trophoblasts in culture. TSCs are prepared from mouse uterine scrapings grown in the presence of fibroblast growth factor, activin, and heparin. TSCs spontaneously differentiate into giant cells when these factors are removed from the medium (FIG. 13). TSCs stained with PrS, whereas differentiated trophoblasts derived from these cells in culture did not stain (FIG. 14). We have also determined that PrS is internalized into stem cells without apoptotic induction. This result confirms observations made in tumor cell lines, in which apoptosis was induced. Without induction of apoptosis, minimal staining was observed in tumor cells. To test for internalization in stem cells, we employed mesenchymal stem cells (MSCs) and TSCs. MSCs were prepared from mouse bone marrow. The bone marrow was flushed from mice and cultured for 6 days in the absence of growth factors. During this incubation, MSCs and hematopoietic stem cells (HSCs) replicate, whereas fibroblasts adhere but do not multiply beyond a few generations. After 6 days, a monolayer is visible. Upon passage by trypsinization, the adherent MSCs are retained, whereas the HSCs, which grow in suspension, are lost. The fibroblasts do not persist due to absence of growth factors and are also not retained. Thus, this simple procedure results in a nearly homogeneous population of MSCs. To confirm the identity of these cells, we treated the cultures separately with dexamethasone and glycerol phosphate (to induce differentiation into osteoblasts) or dexamethasone and indomethacin (to induce differentiation into adipocytes). The results are shown in FIG. 15. In response to the above treatments, differentiated cells showed the appearance of the respective cells. Adipocytes contained large fat vesicles and osteoblasts were dark with distinctive intracellular collagen and mineralization.

[0282] To assess subcellular staining pattern, undifferentiated MSC were stained with PrS and annexin, as well as Hoechst nuclear staining reagent, and observed with confocal microscopy. Results of the observations are shown in FIG. 16. PrS was rapidly internalized. In the case of MSC, about 1 in 20 cells stained with PrS, consistent with previous data, however the precise percentage that stained was not determined. The morphology of MSCs is heterogeneous, and thee cells secrete abundant material into the medium, some of which adheres to the surface of the chamber slide, making resulting in background in some of the images. Nonetheless, the data clearly show internalized PrS, within 5 minutes of addition and annexin on the surface. TSCs were also stained and imaged as was done with the MSCs. The observations confirm internalization into these cells as well, which also occurs within 5 minutes of addition of the protein. The results are shown in FIG. 17. TSCs are morphologically quite variable, and can be multinucleate in the absence of differentiation, as can be seen in the figure. As with the MSCs, these primary cells shed abundant material into the medium, some of which we have established as extracellular vesicles (previous data). This material again makes the imaging difficult. Some EVs stain with annexin and not PrS, and this phenomenon can be seen in TSCs, in FIG. 18. In this image of a cluster of TSCs, vesicles being released by the cells stain with annexin and not PrS, which is internalized. These patterns raise interesting questions regarding the specificity and binding targets of PrS and annexin. The two proteins are both reputed to bind PS. However, the differential binding to EVs as well as distinct subcellular localization patterns suggest that they are not binding in exactly the same manner. Further studies will be required to establish the basis of this distinction, which may prove to be significant. We have also observed PrS staining of the neural progenitor cell line C17.2 (FIG. 19), which is a transformed cell line capable of differentiation in vitro into astrocytes and other neuronal cells. Approximately 5% of these transformed cells stained, although this percentage is an estimate. Remarkably, entry into TSCs occurred even when the cells were chilled to 4 C. (FIG. 20). However, it must be noted that the chamber could not be continually chilled once placed on the microscope. Nevertheless, the temperature could not have risen much within the 5 min time frame of the imaging procedure. This result, while provocative, must clearly be repeated under more controlled conditions. Should the finding be substantiated, the mechanism would have to be very interesting indeed.

[0283] We have succeeded in staining hematopoietic stem cells (HSC) with PrS. Using flow cytometry we determined that HSC stain with PrS, and have observed internalization of PrS in these cells with confocal microscopy. HSC were identified and isolated using fluorescence activated cell sorting (FACS). The cells were identified in bone marrow as lineage-negative, SCA/c-kit positive cells (FIG. 21). These were then stained with FITC-PrS. Two populations of HSC, short-term and long-term, can be identified with the pattern of SLAM marker staining. The SLAM (Signaling Lymphocyte Activation Molecule) markers CD48, CD150, CD229 and CD244 differentially stain HSC with distinct patterns such that SLAM pattern-positive staining is indicative of the ability to both self-renew and differentiate, whereas SLAM pattern-negative HSC can only differentiate. PrS stained a subset of long-term HSC (FIG. 22), and also short-term HSC (FIG. 23). The cells shown are propidium iodide (PI)-negative, meaning that they are all live cells. This result confirms previous experiments demonstrating that a subset of stem cells stains with PrS without the induction of apoptosis.

[0284] We then proceeded to test for internalization of PrS into HSC. This experiment was complicated by many factors. Perhaps the most difficult was the survival in culture of HSC, which die in large numbers in medium overnight. We therefore had to time the experiment such that flow cytometry analysis and confocal microscopy occurred on the same day. Furthermore, the cells are not adherent, making microscopy less than optimal. To make microscopy more efficient, the cells were resuspended in a small drop of medium. Finally, we needed to make sure that the PrS-stained cells analyzed by microscopy were still alive. Many HSC died during the processes of analysis and isolation. Therefore, PI was added and scanned in addition to the Hoescht nuclear stain, and another channel was employed. The presence of PI-bright nuclei indicated dead cells. Despite these difficulties and the complexities of timing, we were able to perform the experiment, and confirmed internalization of PrS into live HSC (FIG. 24). The cells were confirmed as alive by lack of nuclear PI staining. However, some cells were dead or dying as shown in FIG. 25. Despite the complexity and length of the experiment shown, the results show internalization.

[0285] Finally, in FIG. 26, we have performed preliminary toxicity studies on TSC, and determined that at a concentration of 135 g/ml, viability was reduced only by a very minimal extent after 30 min, from 78% to 74%, relative to PBS. Considering that, at this level, 10% of the culture volume was PrS-containing solution, this result confirmed our qualitative observations that PrS is basically non-toxic to stem cells, and the minor toxicity observed could well be due to contaminating contents of the preparation itself. Lower concentrations of PrS showed no effect on viability. The highest level of protein tested was more than 1000 times the concentration used for staining. While full toxicity studies, which were not formally part of this project, will require much more extensive tests, in our hands PrS exhibits very little toxicity.

Summary

[0286] The above results have shown that PrS is rapidly internalized into an array of cells expressing PrS, including stem cells of many types, which suggests that PrS possesses unique characteristics amenable to manipulation toward the goal of developing a therapeutic agent. In addition, the difference in specificity between PrS and annexin such as seen in FIGS. 3 and 7 suggests that binding itself is different between these two proteins. The mere fact that annexin is a tetramer and PrS is a monomer cannot explain these differences and these data suggest that some other component on the cell surface may be involved in PrS binding. The mechanism of binding, specificity, and internalization of PrS, as well as the capability of modular manipulation provide a host of possibilities.

Example 2

[0287] Stem cells are distinct in phenotype from differentiated cells and may express PS non-apoptotically to avoid the induction of immune responses. Stem cells were stained with a GLA domain molecule of the present disclosure comprising a payload of a fluorescent label, without the induction of apoptosis.

[0288] Trophoblast stem cells, (FIG. 14) which differentiate into several types of trophoblasts in the placenta, stained with Protein S, whereas differentiated trophoblasts derived from these cells in culture did not stain. The stain was able to distinguish between in vivo differentiated stems cells and cells differentiated in vitro.

[0289] This data the molecules of the present disclosure may be employed to target cells in vivo or in ex vivo samples.

CONCLUSIONS

[0290] Protein S was shown to bind apoptotic cells, employing a variety of cell types. Background staining of non-apoptotic control cells was minimal. Protein S staining was also compared to that of annexin, using several cell types. Interestingly, Protein S was found to stain apparent apoptotic bodies that did not stain at all with annexin. This result was unexpected; Protein S and annexin both bind PS and were not expected to exhibit differences in specificity. Further evidence of such differences was the failure of 1,000-fold excess annexin to prevent Protein S binding. Again, this result was difficult to understand if the two proteins bind exactly the same molecular target. Using confocal microscopy, subcellular localization studies of Protein S and annexin showed striking differences between these two proteins, even within the same cell. Whereas annexin was largely surface localized, Protein S was almost always found within the cells, in a variety of cell types. This suggests that Protein S may be developed as a biodelivery vehicle. Time course experiments clearly demonstrated that Protein S internalized within 10 minutes of binding. These experiments also provide possible evidence that Protein S binds earlier in the apoptotic process that annexin. The rapid uptake clearly warrants further studies into this interesting mechanism, which may have significant implications for the development of Protein S as a biodelivery agent. Protein S was also labeled with HYNIC conjugated .sup.99mTc, for SPECT imaging of apoptotic tumors and chemically induced liver apoptosis, in comparison to similarly labeled annexin and a modest increase in liver signal of .sup.99mTc Protein S was observed in cyclohexamide-treated animals. In both Listeria and breast cancer models, Cy5 Protein S preferentially localized to apoptotic tissues, exhibiting strong signal to noise in both cases.

[0291] In summary, Protein S has shown specificity for apoptosis in cell culture and in vivo. Remarkably, this protein exhibits distinct specificity and cellular localization compared to annexin with internalization of Protein S, which occurs rapidly an aspect that warrants further exploration.

REFERENCES

[0292] 1. Thompson, W. W., et al., Estimates of US influenza-associated deaths made using four different methods. Influenza Other Respir Viruses, 2009.3(1): p. 37-49. [0293] 2. Osterholm, M. T., et al., Efficacy and effectiveness of influenza vaccines: a systematic review and meta-analysis. Lancet Infect Dis, 2012. 12(1): p. 36-44. [0294] 3. Dixit, R., et al., Emergence of oseltamivir resistance: control and management of influenza before, during and after the pandemic. Infect Disord Drug Targets, 2013. 13(1): p. 34-45. [0295] 4. Jefferson, T., et al., Oseltamivir for influenza in adults and children: systematic review of clinical study reports and summary of regulatory comments. BMJ, 2014. 348: p. g 2545. [0296] 5. Godfrey, C., et al., Delivery is key: lessons learnt from developing splice-switching antisense therapies. EMBO Mol Med, 2017. 9(5): p. 545-557. [0297] 6. Kaczmarek, J. C., P. S. Kowalski, and D. G. Anderson, Advances in the delivery of RNA therapeutics: from concept to clinical reality. Genome Med, 2017.9(1): p. 60. [0298] 7. Ling, H., Non-coding RNAs: Therapeutic Strategies and Delivery Systems. Adv Exp Med Biol, 2016. 937: p. 229-37. [0299] 8. Poon, I. K., et al., Apoptotic cell clearance: basic biology and therapeutic potential. Nat Rev Immunol, 2014. 14(3): p. 166-80. [0300] 9. Birge, R. B., et al., Phosphatidylserine is a global immunosuppressive signal in efferocytosis, infectious disease, and cancer. Cell Death Differ, 2016. 23(6): p. 962-78. [0301] 10. Amara, A. and J. Mercer, Viral apoptotic mimicry. Nat Rev Microbiol, 2015. 13(8): p. 461-9. [0302] 11. Moller-Tank, S. and W. Maury, Phosphatidylserine receptors: enhancers of enveloped virus entry and infection. Virology, 2014. 468-470: p. 565-80. [0303] 12. Ludwig, S., et al., MEK inhibition impairs influenza B virus propagation without emergence of resistant variants. FEBS Lett, 2004. 561(1-3): p. 37-43. [0304] 13. Sharma, R., et al., Detection of phosphatidylserine-positive exosomes for the diagnosis of early-stage malignancies. Br J Cancer, 2017. 117(4): p. 545-552. [0305] 14. Azuma, K., et al., Liver-specific gamma-glutamyl carboxylase-deficient mice display bleeding diathesis and short life span. PLoS One, 2014.9(2): p. e88643. [0306] 15. Hjortoe, G., et al., Factor VIIa binding and internalization in hepatocytes. J Thromb Haemost, 2005. 3(10): p. 2264-73. [0307] 16. Meliopoulos, V. A., et al., Host gene targets for novel influenza therapies elucidated by high-throughput RNA interference screens. FASEB J, 2012. 26(4): p. 1372-86. [0308] 17. Pleschka, S., et al., Influenza virus propagation is impaired by inhibition of the Raf/MEK/ERK signalling cascade. Nat Cell Biol, 2001. 3(3): p. 301-5. [0309] 18. Makkoch, J., et al., Human microRNAs profiling in response to influenza A viruses (subtypes pH1N1, H3N2, and H5N1). Exp Biol Med (Maywood), 2016. 241(4): p. 409-20. [0310] 19. Wolf, S., et al., MicroRNA Regulation of Human Genes Essential for Influenza A (H7N9) Replication. PLoS One, 2016. 11(5): p. e0155104. [0311] 20. Skalickova, S., et al., Perspective of Use of Antiviral Peptides against Influenza Virus. Viruses, 2015. 7(10): p. 5428-42. [0312] 21. Matsubara, T., et al., Sialic acid-mimic peptides as hemagglutinin inhibitors for anti-influenza therapy. J Med Chem, 2010. 53(11): p. 4441-9. [0313] 22. Wunderlich, K., et al., Identification of a PA-binding peptide with inhibitory activity against influenza A and B virus replication. PLoS One, 2009.4(10): p. e7517. [0314] 23. Ozcan, G., et al., Preclinical and clinical development of siRNA-based therapeutics. Adv Drug Deliv Rev, 2015. 87: p. 108-19. [0315] 24. Mack, S., et al., Pseudo-Ligandless Click Chemistry for Oligonucleotide Conjugation. Curr Protoc Chem Biol, 2016. 8(2): p. 83-95. [0316] 25. Paredes, E. and S. R. Das, Click chemistry for rapid labeling and ligation of RNA. Chembiochem, 2011. 12(1): p. 125-31. [0317] 26. Zheng, Y. and P.A. Beal, Synthesis and evaluation of an alkyne-modified ATP analog for enzymatic incorporation into RNA. Bioorg Med Chem Lett, 2016. 26(7): p. 1799-802. [0318] 27. Jain, N., et al., Current ADC Linker Chemistry. Pharm Res, 2015. 32(11): p. 3526-40.