SERUM ALBUMIN BINDERS

Abstract

The present invention relates to amino acid sequences binding to serum albumin. In particular, the present invention relates to improved immunoglobulin single variable domains (also referred to herein as “ISV's” or “ISVD's”), and more in particular improved heavy-chain immunoglobulin single variable domains (also referred to herein as “ISV's” or “ISVD's”) binding to serum albumin, as well as to proteins, polypeptides and other constructs, compounds, molecules or chemical entities that comprise such improved serum albumin binders.

Claims

1.-34. (canceled)

35. Amino acid sequence that is an immunoglobulin single variable domain capable of binding to serum albumin, in which: (i) CDR1 is GFTFSSFGMS (SEQ ID NO:120), CDR2 is SISGSGSDTL (SEQ ID NO:6) and CDR3 is GGSLSR (SEQ ID NO:7), and in which: the amino acid residue at Kabat position 5 is V; the amino acid residue at Kabat position 11 is V; the amino acid residue at Kabat position 16 is G or N; the amino acid residue at Kabat position 45 is P or L; the amino acid residues at Kabat positions 74 to 76 form an SKN or AKT motif; the amino acid residue at Kabat position 89 is L, A or T; and the amino acid residue at Kabat position 104 is G or T; which amino acid sequence has no more than 7 amino acid differences with the sequence of SEQ ID NO:1, wherein the amino acid sequences of the CDRs, the amino acids at Kabat positions 5, 11, 16, 45, 74 to 76, 89 and 104, and any C-terminal extension are not taken into account in determining the number of amino acid differences.

36. Amino acid sequence according to claim 35, in which: the amino acid residue at position 5 is V; the amino acid residue at position 11 is V; the amino acid residue at position 16 is N; the amino acid residue at position 45 is L; the amino acid residues at positions 74 to 76 form an AKT motif; the amino acid residue at position 89 is L, A or T; and the amino acid residue at position 104 is G or T.

37. Protein, polypeptide or other construct, compound, molecule or chemical entity that comprises at least one amino acid sequence according to claim 35.

38. Protein, polypeptide or other construct, compound, molecule or chemical entity according to claim 37, that comprises at least one therapeutic moiety or entity.

39. (canceled)

40. Pharmaceutical composition comprising a protein, polypeptide or other construct, compound, molecule or chemical entity according to claim 37.

41. Amino acid sequence according to claim 35, in which the amino acid at Kabat position 16 is G.

42. Amino acid sequence according to claim 35, in which the amino acid at Kabat position 16 is N.

43. Amino acid sequence according to claim 35, in which the amino acid at Kabat position 45 is P.

44. Amino acid sequence according to claim 35, in which the amino acid at Kabat position 45 is L.

45. Amino acid sequence according to claim 35, in which the amino acids at Kabat positions 74 to 76 form an SKN motif.

46. Amino acid sequence according to claim 35, in which the amino acids at Kabat positions 74 to 76 form an AKT motif.

47. Amino acid sequence according to claim 35, in which the amino acid at Kabat position 89 is L.

48. Amino acid sequence according to claim 35, in which the amino acid at Kabat position 89 is A.

49. Amino acid sequence according to claim 35, in which the amino acid at Kabat position 89 is T.

50. Amino acid sequence according to claim 36, in which the amino acid at Kabat position 89 is L.

51. Amino acid sequence according to claim 36, in which the amino acid at Kabat position 89 is A.

52. Amino acid sequence according to claim 36, in which the amino acid at Kabat position 89 is T.

53. Nucleic acid that encodes the amino acid of claim 35.

54. Method for preparing the amino acid sequence of claim 35, which method comprises cultivating or maintaining a host cell under conditions such that said host cell produces or expresses the amino acid sequence, optionally further isolating the amino acid sequence so produced.

Description

[0639] The invention will now be further described by means of the following non-limiting preferred aspects, examples and figures, in which:

[0640] FIG. 1 is a table listing some of the amino acid positions that will be specifically referred to herein and their numbering according to some alternative numbering systems (such as Aho and IMGT);

[0641] FIG. 2 lists the amino acid sequences referred to herein;

[0642] FIGS. 3A-3C shows alignments of SEQ ID NOs: 1, 50 and 119, and FIG. 3D shows binding of pre-existing antibodies from 6 serum-albumin depleted sera;

[0643] FIG. 4A shows an alignment of SEQ ID NOs: 1 and 8-50 and FIG. 4B shows an alignment of SEQ ID NOs: 1, 50 and 61-102;

[0644] FIG. 5 shows two corresponding plots of data points obtained in Example 1 when 96 serum samples (68 from human healthy subjects, 17 from healthy human volunteers whose serum contained pre-existing antibodies that are capable of binding even in the presence of a C-terminal alanine, and 11 from SLE patients) were tested for binding to [Reference A], [Reference A+C-terminal alanine], and three variants of the invention (i.e. [Reference A+L11V+V89T+C-terminal alanine], [Reference A+L11V+V89T+T110K+C-terminal alanine] and [Reference A+L11V+V89T+S104T+C-terminal alanine], respectively). It should be noted that compared to the prior art sequences of SEQ ID NO:1 and 50, Reference A already contains an L5V mutation, so that all three variants contain V's at positions 5 and 11 in addition to the specific mutations indicated). Each dot represents the binding level for one of the 96 samples tested. The data points shown in the right hand panel and the left hand panel are the same; in the right hand panel the data points measured with each individual sample for each of the compounds tested are connected by means of a line (as a result, the declination of the line gives an indication of the extent to which binding by pre-existing antibodies is reduced when the mutations of the invention and the C-terminal alanine are introduced);

[0645] FIG. 6 shows in detail the data from FIG. 5 that was obtained in Example 1 for the 11 samples from SLE patients;

[0646] FIG. 7 is a table listing the binding data of the data points compiled in FIG. 5;

[0647] FIG. 8 shows two corresponding plots of data points obtained in Example 2 when 96 serum samples (68 from human healthy subjects, 17 from healthy human volunteers whose serum contained pre-existing antibodies that are capable of binding even in the presence of a C-terminal alanine, and 11 from SLE patients) were tested for binding to [Reference A], [Reference A+C-terminal alanine], and three variants of the invention (i.e. [Reference A+L11V+V89A+C-terminal alanine], [Reference A+L11V+V89A+T110K+C-terminal alanine] and [Reference A+L11V+V89L+S104G+C-terminal alanine], respectively). It should be noted that compared to the prior art sequences of SEQ ID NO:1 and 50, Reference A already contains an L5V mutation, so that all three variants contain V's at positions 5 and 11 in addition to the specific mutations indicated). Each dot represents the binding level for one of the 96 samples tested. The data points shown in the right hand panel and the left hand panel are the same; in the right hand panel the data points measured with each individual sample for each of the compounds tested are connected by means of a line (as a result, the declination of the line gives an indication of the extent to which binding by pre-existing antibodies is reduced when the mutations of the invention and the C-terminal alanine are introduced);

[0648] FIG. 9 shows in detail the data from FIG. 8 that was obtained in Example 2 for the 11 samples from SLE patients;

[0649] FIG. 10 is a table listing the binding data of the data points compiled in FIG. 8;

[0650] FIG. 11 shows two corresponding plots of data points obtained in Example 3 when 96 serum samples (68 from human healthy subjects, 17 from healthy human volunteers whose serum contained pre-existing antibodies that are capable of binding even in the presence of a C-terminal alanine, and 11 from SLE patients) were tested for binding to [Reference A], [Reference A+C-terminal alanine], and three variants of the invention (i.e. [Reference A+L11V+V89L+S101G+C-terminal alanine], [Reference A+L11V+V89L+S104A+C-terminal alanine] and [Reference A+L11V+V89L+S101E+C-terminal alanine], respectively). It should be noted that compared to the prior art sequences of SEQ ID NO:1 and 50, Reference A already contains an L5V mutation, so that all three variants contain V's at positions 5 and 11 in addition to the specific mutations indicated). Each dot represents the binding level for one of the 96 samples tested. The data points shown in the right hand panel and the left hand panel are the same; in the right hand panel the data points measured with each individual sample for each of the compounds tested are connected by means of a line (as a result, the declination of the line gives an indication of the extent to which binding by pre-existing antibodies is reduced when the mutations of the invention and the C-terminal alanine are introduced);

[0651] FIG. 12 shows in detail the data from FIG. 11 that was obtained in Example 3 for the 11 samples from SLE patients;

[0652] FIG. 13 is a table listing the binding data of the data points compiled in FIG. 11;

[0653] FIG. 14 shows two corresponding plots of data points obtained in Example 4 when 96 serum samples (68 from human healthy subjects, 17 from healthy human volunteers whose serum contained pre-existing antibodies that are capable of binding even in the presence of a C-terminal alanine, and 11 from SLE patients) were tested for binding to [Reference A], [Reference A+C-terminal alanine], and three variants of the invention (i.e. [Reference A+L11V+R30T+V89L+C-terminal alanine], [Reference A+L11V+S31D+V89L+C-terminal alanine] and [Reference A+L11V+V89S+C-terminal alanine], respectively). It should be noted that compared to the prior art sequences of SEQ ID NO:1 and 50, Reference A already contains an L5V mutation, so that all three variants contain V's at positions 5 and 11 in addition to the specific mutations indicated). Each dot represents the binding level for one of the 96 samples tested. The data points shown in the right hand panel and the left hand panel are the same; in the right hand panel the data points measured with each individual sample for each of the compounds tested are connected by means of a line (as a result, the declination of the line gives an indication of the extent to which binding by pre-existing antibodies is reduced when the mutations of the invention and the C-terminal alanine are introduced);

[0654] FIG. 15 shows in detail the data from FIG. 14 that was obtained in Example 4 for the 11 samples from SLE patients;

[0655] FIG. 16 is a table listing the binding data of the data points compiled in FIG. 14;

[0656] FIG. 17 shows two corresponding plots of data points obtained in Example 5 when 96 serum samples (69 from human healthy subjects, 17 from healthy human volunteers whose serum contained pre-existing antibodies that are capable of binding even in the presence of a C-terminal alanine, and 10 from SLE patients) were tested for binding to [Reference A], [Reference A+C-terminal alanine], and three variants of the invention (i.e. [Reference A+L11V+V89N+C-terminal alanine], [Reference A+L11V+V89N+T110K+C-terminal alanine] and [Reference A+L11V+V89S+T110K+C-terminal alanine], respectively). It should be noted that compared to the prior art sequences of SEQ ID NO:1 and 50, Reference A already contains an L5V mutation, so that all three variants contain V's at positions 5 and 11 in addition to the specific mutations indicated). Each dot represents the binding level for one of the 96 samples tested. The data points shown in the right hand panel and the left hand panel are the same; in the right hand panel the data points measured with each individual sample for each of the compounds tested are connected by means of a line (as a result, the declination of the line gives an indication of the extent to which binding by pre-existing antibodies is reduced when the mutations of the invention and the C-terminal alanine are introduced);

[0657] FIG. 18 shows in detail the data from FIG. 17 that was obtained in Example 5 for the 10 samples from SLE patients;

[0658] FIG. 19 is a table listing the binding data of the data points compiled in FIG. 17;

[0659] FIG. 20 shows two corresponding plots of data points obtained in Example 6 when 96 serum samples (69 from human healthy subjects, 17 from healthy human volunteers whose serum contained pre-existing antibodies that are capable of binding even in the presence of a C-terminal alanine, and 10 from SLE patients) were tested for binding to [Reference A], [Reference A+C-terminal alanine], and three variants of the invention (i.e. [Reference A+L11V+V89L+S101H+C-terminal alanine], [Reference A+L11V+V89L+R102D+C-terminal alanine] and [Reference A+L11V+C-terminal alanine], respectively). It should be noted that compared to the prior art sequences of SEQ ID NO:1 and 50, Reference A already contains an L5V mutation, so that all three variants contain V's at positions 5 and 11 in addition to the specific mutations indicated). Each dot represents the binding level for one of the 96 samples tested. The data points shown in the right hand panel and the left hand panel are the same; in the right hand panel the data points measured with each individual sample for each of the compounds tested are connected by means of a line (as a result, the declination of the line gives an indication of the extent to which binding by pre-existing antibodies is reduced when the mutations of the invention and the C-terminal alanine are introduced);

[0660] FIG. 21 shows in detail the data from FIG. 20 that was obtained in Example 6 for the 10 samples from SLE patients;

[0661] FIG. 22 is a table listing the binding data of the data points compiled in FIG. 20.

[0662] FIG. 23 shows an alignment of the sequences of SEQ ID NOs: 1, 50, 119 and 121 to 132.

[0663] FIGS. 24A to C are alignments of the sequences of SEQ ID NOs: 145 to 184, SEQ ID NOs:185 to 208 and SEQ ID NOs: 209 to 244, respectively, in each case aligned with the sequences of SEQ ID NOs: 1, 50 and 119.

[0664] FIG. 25 shows binding data for some representative albumin binders with an S, an R or a T at position 30.

EXPERIMENTAL PART

[0665] The human samples used in the Experimental Part below were either obtained from commercial sources or from human volunteers (after all required consents and approvals were obtained) and were used in according with the applicable legal and regulatory requirements (including but not limited to those regarding medical secret and patient privacy)

[0666] In the Examples below, unless explicitly indicated otherwise, the binding of pre-existing antibodies that are present in the samples used (i.e. from healthy volunteers, rheumatoid arthritis (RA) patients and SLE patients) to the Nanobodies tested was determined using ProteOn as follows:

[0667] Nanobodies were captured either on serum albumin or via a FLAGS tag using monoclonal anti-FLAG M2.

[0668] In case of binding of pre-existing antibodies on Nanobodies captured on human serum albumin (HSA) was evaluated using the ProteOn XPR36 (Bio-Rad Laboratories, Inc.). PBS/Tween (phosphate buffered saline, pH7.4, 0.005% Tween20) was used as running buffer and the experiments were performed at 25° C. The ligand lanes of a ProteOn GLC Sensor Chip were activated with EDC/NHS (flow rate 30 μl/min) and HSA was injected at 10 μg/ml in ProteOn Acetate buffer pH4.5 (flow rate 100 μl/min) to render immobilization levels of approximately 3200 RU. After immobilization, surfaces were deactivated with ethanolamine HCl (flow rate 30 μl/min). Nanobodies were injected for 2 minutes at 45 μl/min over the HSA surface to render a Nanobody capture level of approximately 200 RU. The samples containing pre-existing antibodies were centrifuged for 2 minutes at 14,000 rpm and supernatant was diluted 1:10 in PBS-Tween20 (0.005%) before being injected for 2 minutes at 45 μl/min followed by a subsequent 400 seconds dissociation step. After each cycle (i.e. before a new Nanobody capture and blood sample injection step) the HSA surfaces were regenerated with a 2 minute injection of HCl (100 mM) at 45 μl/min. Sensorgram processing and data analysis was performed with ProteOn Manager 3.1.0 (Bio-Rad Laboratories, Inc.). Sensorgrams showing pre-existing antibody binding were obtained after double referencing by subtracting 1) Nanobody-HSA dissociation and 2) non-specific binding to reference ligand lane. Binding levels of pre-existing antibodies were determined by setting report points at 125 seconds (5 seconds after end of association). Percentage reduction in pre-existing antibody binding was calculated relative to the binding levels at 125 seconds of a reference Nanobody.

[0669] In case of binding of pre-existing antibodies on FLAG-tagged Nanobodies captured on monoclonal anti-FLAG M2 (Sigma) was evaluated using the ProteOn XPR36 (Bio-Rad Laboratories, Inc.). PBS/Tween (phosphate buffered saline, pH7.4, 0.005% Tween20) was used as running buffer and the experiments were performed at 25° C. The ligand lanes of a ProteOn GLC Sensor Chip were activated with EDC/NHS (flow rate 30 μl/min) and anti-FLAG M2 mAb was injected at 10 μg/ml in ProteOn Acetate buffer pH4.5 (flow rate 100 μl/min) to render immobilization levels of approximately 4000 RU. After immobilization, surfaces were deactivated with ethanolamine HCl (flow rate 30 μl/min). Nanobodies were injected for 2 minutes at 45 μl/min over the anti-FLAG M2 surface to render a Nanobody capture level of approximately 100 RU. To reduce non-specific binding of the blood samples to the anti-FLAG M2 surface 100 nM 3xFLAG peptide (Sigma) was added to the blood samples. The samples containing pre-existing antibodies were centrifuged for 2 minutes at 14,000 rpm and supernatant was diluted 1:10 in PBS-Tween20 (0.005%) before being injected for 2 minutes at 45 μl/min followed by a subsequent 600 seconds dissociation step. After each cycle (i.e. before a new Nanobody capture and blood sample injection step) the anti-FLAG M2 surfaces were regenerated with a 10 seconds injection of Glycine pH1.5 (10 mM) at 150 μl/min. Sensorgram processing and data analysis was performed with ProteOn Manager 3.1.0 (Bio-Rad Laboratories, Inc.). Sensorgrams showing pre-existing antibody binding were obtained after double referencing by subtracting 1) Nanobody-anti-FLAG M2 dissociation and 2) non-specific binding to reference ligand lane. Binding levels of pre-existing antibodies were determined by setting report points at 125 seconds (5 seconds after end of association). Percentage reduction in pre-existing antibody binding was calculated relative to the binding levels at 125 seconds of a reference Nanobody.

Example 1

[0670] Reference A (SEQ ID NO: 119), Reference A with a C-terminal alanine, and three variants of the invention (i.e. [Reference A+L11V+V89T+C-terminal alanine], [Reference A+L11V+V89T+T110K+C-terminal alanine] and [Reference A+L11V+V89T+S104T+C-terminal alanine], respectively), all provided with an N-terminal HIS6 tag, were tested for binding by pre-existing antibodies that are present in 96 serum samples (68 from human healthy subjects, 17 from healthy human volunteers whose serum contained pre-existing antibodies that are capable of binding even in the presence of a C-terminal alanine, and 11 from SLE patients). The compounds were captured using immobilized human serum albumin and binding was measured using ProteOn according to the protocol given in the preamble to this Experimental Part.

[0671] The results are shown in FIGS. 5 (all samples) and 6 (SLE samples only). FIG. 7 lists the results for each of the samples that forms one of the data points in FIG. 5.

[0672] Also, for the serum albumin binders tested, a kinetic analysis was performed of the binding interaction with the immobilized serum albumin (Langmuir, simultaneous ka/kd model). The results are listed in Table C.

TABLE-US-00003 TABLE C ka kd KD Nanobody (1/Ms) T(ka) (1/s) T(kd) (M) Ref. A 2.90E+05 110 2.20E−03 108 7.60E−09 Ref. A + Ala 3.30E+05 124 2.30E−03 121 6.90E−09 Ref. A + L11V + 3.10E+05 112 2.30E−03 112 7.30E−09 V89T + Ala Ref. A + L11V + 2.80E+05 393 2.10E−03 173 7.60E−09 V89T + T110K + Ala Ref. A + L11V + 2.80E+05 90 3.30E−03 126 1.20E−08 V89T + S104T + Ala

Example 2

[0673] Reference A (SEQ ID NO: 119), Reference A with a C-terminal alanine, and three variants of the invention (i.e. [Reference A+L11V+V89A+C-terminal alanine], [Reference A+L11V+V89A+T110K+C-terminal alanine] and [Reference A+L11V+V89L+S104G+C-terminal alanine], respectively), all provided with an N-terminal HIS6 tag, were tested for binding by pre-existing antibodies that are present in 96 serum samples (68 from human healthy subjects, 17 from healthy human volunteers whose serum contained pre-existing antibodies that are capable of binding even in the presence of a C-terminal alanine, and 11 from SLE patients). The compounds were captured using immobilized human serum albumin and binding was measured using ProteOn according to the protocol given in the preamble to this Experimental Part.

[0674] The results are shown in FIGS. 8 (all samples) and 9 (SLE samples only). FIG. 10 lists the results for each of the samples that forms one of the data points in FIG. 8.

[0675] Also, for the serum albumin binders tested, a kinetic analysis was performed of the binding interaction with the immobilized serum albumin (Langmuir, simultaneous ka/kd model). The results are listed in Table D.

TABLE-US-00004 TABLE D ka kd KD Nanobody (1/Ms) T(ka) (1/s) T(kd) (M) Ref. A 2.5E+05 165 1.9E−03 218 7.6E−09 Ref. A + Ala 3.2E+05 175 2.0E−03 206 6.4E−09 Ref. A + L11V + 2.6E+05 84 1.9E−03 106 7.3E−09 V89A + Ala Ref.A + L11V + 2.7E+05 134 1.8E−03 156 6.5E−09 V89A + T110K + Ala Ref. A + L11V + 2.6E+05 161 2.2E−03 207 8.3E−09 V89L + S104G + Ala

Example 3

[0676] Reference A (SEQ ID NO: 119), Reference A with a C-terminal alanine, and three variants of the invention (i.e. [Reference A+L11V+V89L+S101G+C-terminal alanine], [Reference A+L11V+V89L+S104A+C-terminal alanine] and [Reference A+L11V+V89L+S101E+C-terminal alanine], respectively), all provided with an N-terminal HIS6 tag, were tested for binding by pre-existing antibodies that are present in 96 serum samples (68 from human healthy subjects, 17 from healthy human volunteers whose serum contained pre-existing antibodies that are capable of binding even in the presence of a C-terminal alanine, and 11 from SLE patients). The compounds were captured using immobilized human serum albumin and binding was measured using ProteOn according to the protocol given in the preamble to this Experimental Part.

[0677] The results are shown in FIGS. 11 (all samples) and 12 (SLE samples only). FIG. 13 lists the results for each of the samples that forms one of the data points in FIG. 11.

[0678] Also, for the serum albumin binders tested, a kinetic analysis was performed of the binding interaction with the immobilized serum albumin (Langmuir, simultaneous ka/kd model). The results are listed in Table E.

TABLE-US-00005 TABLE E ka kd KD Nanobody (1/Ms) T(ka) (1/s) T(kd) (M) Ref. A 1.1E+05 88 1.6E−03 193 1.4E−08 Ref. A + Ala 7.5E+04 59 1.5E−03 185 2.0E−08 Ref. A + L11V + 1.1E+05 131 1.7E−03 279 1.6E−08 V89L + S101G + Ala Ref. A + L11V+ 7.2E+04 71 2.2E−03 268 3.0E−08 V89L + S104A + Ala Ref. A + L11V + 1.5E+05 128 4.4E−03 391 3.0E−08 V89L + S101E + Ala

Example 4

[0679] Reference A (SEQ ID NO: 119), Reference A with a C-terminal alanine, and three variants of the invention (i.e. [Reference A+L11V+R30T+V89L+C-terminal alanine], [Reference A+L11V+S31D+V89L+C-terminal alanine] and [Reference A+L11V+V89S+C-terminal alanine], respectively), all provided with an N-terminal HIS6 tag, were tested for binding by pre-existing antibodies that are present in 96 serum samples (68 from human healthy subjects, 17 from healthy human volunteers whose serum contained pre-existing antibodies that are capable of binding even in the presence of a C-terminal alanine, and 11 from SLE patients). The compounds were captured using immobilized human serum albumin and binding was measured using ProteOn according to the protocol given in the preamble to this Experimental Part.

[0680] The results are shown in FIGS. 14 (all samples) and 15 (SLE samples only). FIG. 16 lists the results for each of the samples that forms one of the data points in FIG. 14.

[0681] Also, for the serum albumin binders tested, a kinetic analysis was performed of the binding interaction with the immobilized serum albumin (Langmuir, simultaneous ka/kd model). The results are listed in Table F.

TABLE-US-00006 TABLE F ka kd KD Nanobody (1/Ms) T(ka) (1/s) T(kd) (M) Ref. A 1.60E+05 147 1.70E−03 247 1.10E−08 Ref. A + Ala 2.10E+05 1259 1.70E−03 331 8.00E−09 Ref. A + L11V + 1.20E+05 109 2.60E−03 351 2.20E−08 R30T + V89L + Ala Ref. A + L11V + 1.60E+05 122 3.20E−03 364 2.00E−08 S31D + V89L + Ala Ref. A + L11V + 1.60E+05 147 1.70E−03 247 1.10E−08 V89S + Ala

Example 5

[0682] Reference A (SEQ ID NO: 119), Reference A with a C-terminal alanine, and three variants of the invention (i.e. [Reference A+L11V+V89N+C-terminal alanine], [Reference A+L11V+V89N+T110K+C-terminal alanine] and [Reference A+L11V+V89S+T110K+C-terminal alanine], respectively), all provided with an N-terminal HIS6 tag, were tested for binding by pre-existing antibodies that are present in 96 serum samples (69 from human healthy subjects, 17 from healthy human volunteers whose serum contained pre-existing antibodies that are capable of binding even in the presence of a C-terminal alanine, and 10 from SLE patients). The compounds were captured using immobilized human serum albumin and binding was measured using ProteOn according to the protocol given in the preamble to this Experimental Part.

[0683] The results are shown in FIGS. 17 (all samples) and 18 (SLE samples only). FIG. 19 lists the results for each of the samples that forms one of the data points in FIG. 17.

[0684] Also, for the serum albumin binders tested, a kinetic analysis was performed of the binding interaction with the immobilized serum albumin (Langmuir, simultaneous ka/kd model). The results are listed in Table G.

TABLE-US-00007 TABLE G ka kd KD Nanobody (1/Ms) T(ka) (1/s) T(kd) (M) Ref. A 1.9E+05 146 1.5E−03 199 8.1E−09 Ref. A + Ala 2.0E+05 151 1.7E−03 215 8.5E−09 Ref. A + L11V + 1.2E+06 37 3.6E−03 57 3.1E−09 V89N + Ala Ref. A + L11V + 2.2E+05 179 1.6E−03 223 7.3E−09 V89N + T110K + Ala Ref. A + L11V + 2.0E+05 139 1.5E−03 172 7.2E−09 V89S + T110K + Ala

Example 6

[0685] Reference A (SEQ ID NO: 119), Reference A with a C-terminal alanine, and three variants of the invention (i.e. [Reference A+L11V+V89L+S101H+C-terminal alanine], [Reference A+L11V+V89L+R102D+C-terminal alanine] and [Reference A+L11V+C-terminal alanine], respectively), all provided with an N-terminal HIS6 tag, were tested for binding by pre-existing antibodies that are present in 96 serum samples (69 from human healthy subjects, 17 from healthy human volunteers whose serum contained pre-existing antibodies that are capable of binding even in the presence of a C-terminal alanine, and 10 from SLE patients). The compounds were captured using immobilized human serum albumin and binding was measured using ProteOn according to the protocol given in the preamble to this Experimental Part.

[0686] The results are shown in FIGS. 20 (all samples) and 21 (SLE samples only). FIG. 22 lists the results for each of the samples that forms one of the data points in FIG. 20.

[0687] Also, for the serum albumin binders tested, a kinetic analysis was performed of the binding interaction with the immobilized serum albumin (Langmuir, simultaneous ka/kd model). The results are listed in Table H.

TABLE-US-00008 TABLE H ka kd KD Nanobody (1/Ms) T(ka) (1/s) T(kd) (M) Ref. A 1.10E+05 107 1.60E−03 251 1.50E−08 Ref. A + Ala 1.70E+05 161 1.60E−03 256 9.80E−09 Ref. A + L11V + 1.10E+05 116 2.60E−03 404 2.40E−08 V89L + S101H + Ala Ref. A + L11V + 6.20E+04 96 5.10E−03 507 8.20E−08 V89L + R102D + Ala Ref. A + L11V 1.50E+05 142 1.70E−03 243 1.20E−08

Example 7: Influence of Amino Acid Residue at Position 30 Binding to Serum Albumin

[0688] The kinetic binding data (on-rate, off-rate and affinity) obtained for some representative serum albumin binders (all having a V at position 5 and a V at position 11) for binding to guinea pig serum albumin, rat serum albumin, mouse serum albumin, cyno serum albumin and human serum albumin, respectively, was determined using ProteOn. The sequence of SEQ ID NO:245 was used as a reference. The results are shown in FIG. 25 and show that the tested albumin binders with an S, T and R at position 30, respectively, had comparable affinities (expressed as KD values) for the different mammalian serum albumins used (i.e. in comparable in respect of binding to guinea pig serum albumin, comparable in respect of binding to rat serum albumin, etc.).

[0689] The entire contents of all of the references (including literature references, issued patents, published patent applications, and co-pending patent applications) cited throughout this application are hereby expressly incorporated by reference, in particular for the teaching that is referenced hereinabove.