HMGB1 tyrosine mutants

Abstract

The present invention relates to a high mobility group box 1 (HMGB1) polypeptide, wherein in said HMGB1 polypeptide at least one, preferably two, more preferably three, most preferably all four tyrosine residues at positions corresponding to amino acid positions Y109, Y144, Y155 and/or Y162 of human HMGB1 have been exchanged to an amino acid residue independently selected from glutamic acid, glutamine, aspartic acid, asparagine, homoglutamic acid (2-aminohexanedioic acid), and homoglutamine (2,6-diamino-6-oxohexanoic acid). The present invention further relates to a polynucleotide encoding a polypeptide according to the present invention, to a vector comprising said polynucleotide, and to a host cell comprising said polypeptide, said polynucleotide and/or said vector. Also, the present invention relates to methods, kits, and uses related thereto.

Claims

1. A high mobility group box 1 (HMGB1) polypeptide, wherein in said HMGB1 polypeptide, at least one tyrosine residue at a position corresponding to one of amino acid positions 22, 57, 68, and 75 of SEQ ID NO: 1 has been exchanged to an amino acid residue independently selected from glutamic acid, glutamine, aspartic acid, asparagine, homoglutamic acid (2-aminohexanedioic acid), and homoglutamine (2,6-diamino-6-oxohexanoic acid), and wherein said HMGB1 polypeptide otherwise comprises an amino acid sequence having at least 95%, at least 98% or 100% sequence identity with the amino acid sequence of SEQ ID NO: 3.

2. The HMGB1 polypeptide of claim 1, wherein in said HMGB1 polypeptide at least two tyrosine residues corresponding to said positions have been exchanged to amino acid residues independently selected from glutamic acid, glutamine, aspartic acid, and asparagine.

3. The HMGB1 polypeptide of claim 1, wherein in said HMGB1 polypeptide all four tyrosine residues at said positions have been exchanged to glutamic acid residues or have been exchanged to glutamine residues.

4. The HMGB1 polypeptide of claim 1, wherein in said HMGB1 polypeptide all four tyrosine residues at said positions have been exchanged to glutamic acid residues.

5. The HMGB1 polypeptide of claim 1, wherein in said HMGB1 polypeptide all four tyrosine residues at said positions have been exchanged to glutamine residues.

6. The HMGB1 polypeptide of claim 1, wherein said HMGB1 polypeptide comprises an amino acid sequence having at least 95% sequence identity to any one of SEQ ID NOS: 5 to 8.

7. The HMGB1 polypeptide of claim 1, wherein said HMGB1 polypeptide is derived from a human HMGB1 polypeptide.

8. The HMGB1 polypeptide of claim 1, wherein said HMGB1 polypeptide induces increased cell death in cultured SW480 cells at a concentration of 0.8 μM compared to a polypeptide comprising an amino acid sequence of SEQ ID NO:3.

9. A polynucleotide encoding a polypeptide according to claim 1.

10. A method for treating cancer and/or for inducing immune modulation in a subject comprising a) contacting said subject with a polypeptide according to claim 1, and, thereby b) treating cancer and/or inducing immune modulation in said subject.

11. The method of claim 10, wherein said treating cancer and/or said immune modulation comprises cellular cancer immune therapy.

12. The method of claim 10, wherein said treating cancer and/or said immune modulation comprises inducing TNF-alpha secretion by NK cells.

13. The method of claim 10, wherein said treating cancer and/or said immune modulation comprises avoiding inducing an adverse event.

14. The method of claim 10, wherein said treating cancer and/or said immune modulation comprises avoiding inducing cytokine release syndrome (CRS).

Description

FIGURE LEGENDS

(1) FIG. 1: Potent activation of Natural Killer cells (NK-92 CI) by Glu-HMGB1 (n=3, p<0.05). WT=HMGB1 WT, E-Variant=Glu-HMGB1 (SEQ ID NO: 8), Q-Variant=Gln-HMGB1 (SEQ ID NO: 6).

(2) The following Examples shall merely illustrate the invention. They shall not be construed, whatsoever, to limit the scope of the invention.

EXAMPLE 1: METHODS

(3) Cr-51 Release Assay

(4) SW480 colon cancer cells were cultured in 96-well plates (20000 cells/well) and labeled with .sup.51Cr (25 μCi/well) for 2 h. Then cells were treated with HMGB1 and NK-cells for 24 h hours. Medium was removed and counted for radioactivity (cpm1). Cells were washed 4× with medium and solubilized with 100 μl 0.5N NaOH. The lysates were counted for radioactivity (cpm2).

(5) Radioactivity Counting

(6) Samples were mixed with 10 ml UltimaGold and counted in a liquid scintillation counter.

(7) $\mathrm{Calculation}\mathrm{of}\mathrm{cell}\mathrm{death}\text{:}\frac{\mathrm{cpm}1}{\mathrm{cpm}1+\mathrm{cpm}2}\times 100=\%\mathrm{cell}\mathrm{death}.$
Generation of GluHMGB1, GlnHMGB1 and Wildtype HMGB1

(8) Plasmids encoding a HMGB1 polypeptide with all four B-Box domain tyrosine residues replaced by glutamate or glutamine residues, respectively, were transfected into HEK cells (serum-free suspension cell culture, 1,000 ml (app. 2.5×10.sup.6 cells/ml), then supplemented with Valproic Acid). For generation of wildtype HMGB1, the B-Box domain was unmodified on its tyrosine residues. The cell pellet was homogenized and purified via IMAC and TALON (Clontech) Resins and eluted using imidazole. 15 eluates were analyzed via SDS-PAGE (Coomassie staining). After pooling of positive eluates the protein was gel filtrated (Superdex) and finally analyzed by SDS-PAGE. 800 nM of the purified protein was used in the experiments.

(9) Activation of NK Cells

(10) Activation of Natural Killer cell line NK-92 CI was measured by detection of TNF-alpha release into the supernatant using 400,000 NK cells stimulated with 800 nM of the indicated HMGB1 protein for 24 h. Detection of TNF-alpha was performed with an ELISA-Kit (Avia Systems Biology, Catalog No. OKAA00027_96 W) according to the instructions of the manufacturer.

EXAMPLE 2: RESULTS AND DISCUSSION

(11) Different HMGB1 Forms Display Both Distinct Cytotoxicity Towards Cancer Cells and Differences in Enhancement of Natural Killer Cytotoxicity Towards Cancer Cells and Activation of NK Cells

(12) Using both a very short time period for assessment of HMGB1 cytotoxicty of 24 h and the highly sensitive .sup.51Cr (chromium 51) release goldstandard assay for detection of cell death, we revealed surprising new cytotoxicity profiles of different HMGB1 forms, modified on their B-Box domain. Gln-HMGB1 showed 43% more cell death towards cancer cells compared to wildtype HMGB1 whereas Glu-HMGB1 showed even 83% more cell death compared to wildtype HMGB1 (Table 1).

(13) Next, effector (E) cells (NK-92 CI Natural Killer cell line) were incubated with .sup.51Cr labeled target (T) cells (SW480 colon cancer cells) at various E:T ratios: 5:1, 10:1 and 20:1 (Table 1). Gln-HMGB1 plus NK cells showed up to 55% more cell death towards cancer cells compared to NK cell lytic activity with no stimulation whereas Glu-HMGB1 showed up to 94% more cell death towards cancer cells compared to NK cell lytic activity with no stimulation (E:T 5:1 respectively) (Table 1). Further, Gln-HMGB1 plus NK cells showed up to 13% more cell death towards cancer cells compared to NK cell lytic activity after stimulation with wildtype HMGB1 whereas Glu-HMGB1 showed up to 25% more cell death towards cancer cells compared to NK cell lytic activity after stimulation with wildtype HMGB1 (E:T 5:1 respectively) (Table 1).

(14) In summary the three different HMGB1 isoforms display distinct cytotoxicity profiles. Glu-HMGB1 is superior to both wildtype HMGB1 and Gln-HMGB1 regarding cytotoxicity towards cancer cells and regarding enhancement of NK cell lytic activity towards cancer cells. Gln-HMGB1 is superior to wildtype HMGB1 regarding cytotoxicity towards cancer cells and regarding enhancement of NK cell lytic activity towards cancer cells. Thus wildtype HMGB1 displays the poorest cytotoxicity profile compared to Glu-HMGB1 and Gln-HMGB1. Gln-HMGB1 still exerts strong cytotoxicity towards cancer cells (albeit less than Glu-HMGB1) with a moderate enhancement of NK cell lytic activity (ranging in between wildtype HMGB1 (low NK cell lytic activity) and Glu-HMGB1 (high NK cell lytic activity)).

(15) Moreover, the three different HMGB1 isoforms display distinct activities in activating NK cells. We measured the secretion of tumor necrosis factor-alpha (TNF-alpha) by NK-92 CI cells following activation by HMGB1 stimulation. Glu-HMGB1 most potently stimulated secretion of TNF-alpha and thus most potently activated NK cells to become binders and killers.

(16) Thus by the application of different forms of HMGB1, Glu-HMGB1 and Gln-HMGB1, in cancer therapy it is possible to regulate (i) direct cytotoxicity towards cancer cells, (ii) NK cell cytotoxicity towards cancer cells that are in the vicinity of the tumor and (iii) overall activation of circulating NK cells (switching them to binders and killers) and thus facilitating recruitment of NK cells to the tumor. The results shown herein demonstrate that Glu-HMGB1 and Gln-HMGB1 on its own or combined with cellular immune therapies (e.g. NK cells) display augmented cytotoxicity towards cancer cells compared to wildtype HMGB1.

(17) With GluHMGB1 and GlnHMGB1 the clinician has a tool by which he can regulate the general activation of the immune system (e.g. activation of NK cells secreting different cytokines, e.g. TNF-alpha) during HMGB1 based cancer therapy. Depending on the desired activation status or desired cytokine blood plasma levels (e.g. TNF-alpha), Glu-HMGB1 could be used for strong activation of the immune system and enhanced cytokine release while Gln-HMGB1 could be used for a moderate activation of the immune system and moderate cytokine release (e.g. from NK cells). By combining both, Glu-HMGB1 and Gln-HMGB1, one could enhance or decrease the release of cytokines while always having increased direct cytotoxicity towards cancer cells, and NK cell cytotoxicity towards cancer cells that are in the vicinity of the tumor, compared to wildtype HMGB1.

(18) TABLE-US-00001 TABLE 1 Cr-51 release assay using NK-92 CI Natural Killer cells and SW480 colon cancer cells plus HMGB1 (n = 3, *p < 0.05). % % In- % In- In- creased creased creased cell cell cell death death % Cell death com- com- death % Cell com- pared to pared to (mean, death pared only NK WT plus NK cells HMGB1 N = 3) (SD) to WT cells NK cells 0 Control 0.45 0.21 100000 Control 8.05 1.06 0 200000 Control 28.29 1.08 0 400000 Control 56.41 2.27 0 0 HMGB1_WT 3.95 0.35 0 0 HMGB1_Q 5.67 0.57 43* 0 HMGB1_E 7.23 0.21 83* 100000 HMGB1_WT 10.98 0.04 36* 0 100000 HMGB1_Q 12.45 0.71 55* 13* 100000 HMGB1_E 15.59 0.21 94* 25* 200000 HMGB1_WT 31.87 1.36 13* 0 200000 HMGB1_Q 35.56 0.20 26* 12* 200000 HMGB1_E 38.48 1.38 36* 21* 400000 HMGB1_WT 58.81 1.73  4* 0 400000 HMGB1_Q 62.55 0.54 11*  6* 400000 HMGB1_E 69.21 3.54 23* 18*