Methods of Selecting and Detecting Binding Peptides

Abstract

Methods for selecting, detecting, and/or enriching peptides that bind a target are provided.

Claims

1. A method of detecting or selecting for a binding peptide to a target, wherein the method comprises detecting or selecting for an mRNA-DNA-peptide complex attached to at least one detectable substance.

2. The method of claim 1, wherein the detectable substance is a fluorophore, is conjugated to a puromycin linker, or is a tag.

3. The method of claim 2, wherein the fluorophore is an organic dye, a biological fluorophore, or quantum dots.

4. The method of claim 2, wherein the detectable substance is an Alexa Fluor.

5. The method of claim 2, wherein the tag is a peptide tag, or a FLAG™ tag, or a natural or non-natural amino acid.

6. The method of claim 5, wherein the tag is detected by a fluorescent molecule.

7. The method of claim 6, wherein the fluorescent molecule is a fluorescent peptide, or a fluorescent antibody.

8. The method of claim 1, wherein the target is immobilized on material that enables sort.

9. The method of claim 1, wherein the target is a peptide, protein, nucleic acid, sugar, lipid, mammalian cell, or mammalian cell extract.

10. The method of claim 9, wherein the target is a peptide or protein.

11. The method of claim 1, wherein the binding peptide comprises at least one non-natural amino acid.

12. The method of claim 1, wherein the binding peptide is detected or selected for by the detectable substance.

13. The method of claim 1, wherein the binding peptide is detected or selected for by FACS, flow cytometry, ELISA, spectrophotometry, fluorescence spectroscopy, or microscopy.

14. The method of claim 1, wherein the binding peptide is detected and selected.

15. The method of claim 1, wherein the method comprises: a) incorporating at least one detectable substance into an mRNA library to produce an mRNA-detectable substance complex; b) translating the mRNA-detectable substance complex to produce an mRNA-peptide-detectable substance complex; and c) reverse transcribing the mRNA-peptide-detectable substance complex to obtain an mRNA-DNA-peptide-detectable substance complex.

16. The method of claim 15, wherein the mRNA-DNA-peptide-detectable substance complex is detected by flow cytometry.

17. The method of claim 15, wherein the method further comprises transcribing a template DNA library to obtain an mRNA library.

18. The method of claim 15, further comprising removing mRNA-DNA-peptide complex not bound to a target.

19. The method of claim 18, wherein the mRNA-DNA-peptide complex is removed by heat.

20. The method of claim 19, further comprising amplifying by PCR the DNA of the removed mRNA-DNA-peptide complex.

21. The method of claim 20, further comprising transcribing the DNA to obtain an mRNA library.

22. The method of claim 1, further comprising determining the amino acid sequence of the binding peptide.

23. The method of claim 1, further comprising determining the nucleic acid sequence of the binding peptide.

24. A library comprising more than one mRNA-DNA-peptide complex of claim 1.

Description

EXAMPLES

Example: Detection of Binding of an mRNA-DNA-Peptide Complex to a Target

[0032] A peptide (herein referred to as “Peptide A”) attached to its mRNA in an mRNA-DNA-peptide detectable complex is used to determine if flow cytometry can be utilized to detect binding of peptide to its target (herein referred to as “Target A”). Target A may be recombinant protein fused to an AVI-tag (a tag that is fused recombinantly to a peptide of interest to facilitate its site-specific biotinylation in vivo by a biotin ligase (BirA) enzyme).

[0033] Messenger RNA encoding each peptide is annealed to a fluorophore (e.g. Alexa Fluor 647)-conjugated puromycin linker. Then, in vitro translation (IVT) is initiated using PDPS and mRNA is reverse transcribed, resulting in an mRNA-DNA-peptide A complex. The mRNA-DNA-peptide A complex is added to biotinylated Target A-Avi tag to allow for binding. Streptavidin-coated beads are added to the IVT mix (containing target A-Avi tag) and the complex of streptavidin-coated beads with biotinylated Target A that is bound to the mRNA-DNA-peptide complex is pulled down, washed, and diluted in FACS buffer prior to performing flow cytometry. Streptavidin beads mixed with in vitro transcription translation mix (IVT mix) that lack amino acids and aminoacyl tRNA synthetase (ARSs) [IVT(−)] is used as a negative control to ensure that binding observed by flow cytometry is specific to Target A-Avi tag binding peptides.

[0034] Following procedures essentially as described above in this Example, it is expected that strong binding of mRNA-DNA-peptide complexes to its target (Target A) will be detected.

[0035] Following procedures essentially as described above in this Example, strong binding of mRNA-DNA-peptide complexes to its target (Target A) was detected.

Example: Target Specific Enrichment During mRNA Library Selection

[0036] To determine if flow cytometry can be used to monitor selection by detecting the enrichment of target-specific peptides, selection rounds against recombinant Target A-Avi conjugated to streptavidin beads can be interrogated.

[0037] Chloroacetyl phenylalanine is used to initiate translation and to make a stable thioether bond with a C-terminally engineered cysteine residue. Selection is monitored by determining percent recovery by qPCR. A negative selection against streptavidin beads is incorporated in each round to eliminate non-specific streptavidin binders. An increase in percent recovery is expected to be most noticeable in the later rounds and would suggest enrichment in the population of identified target A binders.

[0038] Peptide sequences from later rounds are then determined using Illumina® next generation sequencing. A high frequency of a small number of peptides may demonstrate enrichment of those particular peptides. To determine if the observed increase in percent recovery might be due to selection of non-specific binders, peptide sequences from a later round are compared to peptide sequences from a previous round. It is expected that non-specific binders will be enriched at the later round using the qPCR method. These non-specific binders may include peptides with truncated sequences comprised of one or two amino acids.

[0039] Following procedures essentially as described above, a high frequency of a small number of peptides was indicative of enrichment of those particular peptides. Comparison of peptide sequences from a later round to previous round revealed that the highly enriched peptides are truncated sequences comprised of one or two amino acids. In most cases, as the selection rounds increase, non-specific binders will be enriched at the later rounds. qPCR indiscriminately quantifies number of output DNA molecules, therefore true versus non-specific peptides cannot be differentiated using this method. Therefore, other means than qPCR should be employed to improve selection of specific binders.

[0040] To follow enrichment of Target A binding peptides from different rounds of selection, flow cytometry is used. DNA from outputs of different rounds is isolated and amplified by PCR to re-constitute each peptide pool with fluorophore-conjugated puromycin linker, and binding of each pool to Target A-AVI by flow cytometry is assessed. Streptavidin plus IVT(−) sample is used for gating and as negative control. Despite the expected increase in percent recovery by qPCR, it is expected that there will be minimal, if any, increase in binding from the different rounds by flow cytometry. This may be, in part, due to the reduced selection of non-specific binders recovered from flow cytometry.

[0041] Following procedures essentially as described above, an increase in percent recovery by qPCR suggested that later rounds should bind target much stronger. However, specific binding of the peptide pool to the target by flow cytometry was not observed. Lack of binding of peptide pools to target by flow cytometry corroborated with the enrichment of truncated non-specific peptides in later rounds of selection.

Example: Use of FACS to Isolate Target Specific Binders from mRNA Library

[0042] To determine if a library screened against Target A-AVI can be sorted by FACS, DNA from the output of a later round is transcribed and translated using a puromycin linker fused to a fluorophore, such as Alexa Flour 647, and flow cytometry is performed to detect binding. Gating is around Target A-AVI in complex with streptavidin beads, and it is expected that a low percentage of peptides will bind to Target A (pre-sorted pool). The positive population is sorted by flow cytometry, and binding of the peptide pool from the sorted round to Peptide A-AVI is assessed by flow cytometry. The results are expected to indicate a significant increase in binding of the sorted pool compared to the pre-sorted pool. The results are also expected to demonstrate that selection by FACS results in greater enrichment of Target A binding peptides compared to using the traditional mRNA selection method, and that selection by FACS is an efficient method to eliminate background binders and to isolate specific peptides.

[0043] Following procedures essentially as described above, a low percentage of peptides bound to Target A (pre-sorted pool). The positive population was sorted by flow cytometry, and binding of the peptide pool from the sorted round to Target A-AVI was assessed by flow cytometry. The results indicated a significant increase in binding of the sorted pool compared to the pre-sorted pool. The results also demonstrated that selection by FACS resulted in greater enrichment of Target A specific binding peptides compared to using the traditional percent recovery method, and that selection by FACS is an efficient method to eliminate background binders and to isolate specific peptides.

[0044] To further determine if selection by percent recovery can result in enrichment of both specific and non-specific binders, the peptide sequences of a sorted round are compared to peptide sequences of corresponding unsorted rounds. It is expected that several of the highest frequency sequences of a sorted round will be different from unsorted rounds. In addition, the frequency of specific peptides is expected to increase after sorting by FACS. This would suggest that selection of mRNA display libraries by FACS using the modified puromycin linker results in highly efficient isolation of peptides with strong binding affinity and specificity.

[0045] Following procedures essentially as described above, several of the highest frequency sequences of the peptides from sorted rounds are different from unsorted rounds. The frequency of specific peptides was increased after sorting by FACS. This suggests that selection of mRNA display libraries by FACS using the modified puromycin linker results in highly efficient isolation of peptides with strong binding affinity and specificity.

Example: Detection and Selection of Cell-Expressed Peptide

[0046] The conventional mRNA display selection method cannot efficiently detect cell-expressed peptides because of high noise intrinsic to the system resulting in selection of non-specific peptides. Four rounds of selection are completed using PDPS against a recombinant heterodimeric cell expressed protein (herein referred to as “Protein II”; biotinylated; attached to streptavidin beads). Streptavidin beads are used to pull down complexes of peptide with Protein II and selection is monitored by determining percent recovery in each round. A significant increase in percent recovery was observed at round four (Table 1), and therefore the stringency of selection is increased, in part, by decreasing the concentration of target to favor isolation of stronger binders.

TABLE-US-00001 TABLE 1 Percent recovery in each round of selection against recombinant Protein II. % recovery against biotinylated Rounds Protein II 1 0.0024 2 0.0072 3 0.27   4 2.92   5 0.057 

[0047] To determine if a cell-expressed protein can be used as a target by the methods of the present invention, flow cytometry is used against either recombinant or cell-expressed Protein II to monitor enrichment of Protein II-binders during selection. DNA from output of rounds two to five is transcribed, annealed to modified puromycin linker, and translated to constitute a fluorophore library for each round. Binding of each mRNA-DNA-peptide pool to recombinant or cell-expressed Protein II is detected by flow cytometry.

[0048] Following procedures essentially as described above, the outputs of round two to round five bound strongly to both recombinant and cell-expressed Protein II as detected by flow cytometry (Table 2). These data demonstrate that the methods of the present invention can be used to select for both recombinant and cell-expressed peptides.

TABLE-US-00002 TABLE 2 Binding of mRNA-DNA-peptide library to Protein II from different rounds of selection and detected by flow cytometry. % MFI Rounds Recombinant Protein II Cell-Expressed Protein II 2  6.48 25.7  3 77.4  46.1  4 98    53.7  5 99.9  64.3  IVT(-)   0.062  0.91

[0049] The DNA from output of round two is then used to sort Protein II binders using FACS. Since a recombinant form of the protein is used in the first round of selection, sorting is performed with Protein II expressed on CHO cells to isolate macrocycles that bind to both forms (recombinant or cell-expressed) of the target. To determine if sorting by FACS results in enrichment of Protein II binders, binding of pre- and post-sorted peptide pools to Protein II is examined. The results indicate enrichment in the population of specific binders as determined by flow cytometry post-sort (4% binders in pre-sort compared to 40% post-sort). Binding of the sorted pool to parental CHO cells that lack expression of Protein II was minimal Next generation sequencing of round two (sorted) peptides revealed that new families, whose sequences were not present at detectable levels in pre-sort and later rounds of selections (that was carried out by conventional mRNA display selection methods), had emerged.

[0050] To determine how binding and functional activity of the most enriched peptides from the round two (sorted) pool compare to round five, peptides identified from round two (sorted) pool will be chemically synthesized and tested in a cell-based functional assay to determine if binding of Protein II with its ligand can be blocked by the sorted peptides. The data are expected to demonstrate that peptides present in round two (sorted) pool will inhibit the interaction of Protein II with its ligand. These results would indicate that FACS sorting in the earlier rounds of selection results in isolation and identification of peptides with functional activity that might be depleted from the library in later rounds due to carrying out multiple rounds of selection.

[0051] Following procedures essentially as described above, peptides present in round two (sorted) pool inhibited the interaction of Protein II with its ligand.

Example: Design and Optimization of DNA-Linker Conjugated to Fluorophore

[0052] Fluorophores are conjugated (attached) to the 5′-end of puromycin linkers to enable detection of mRNA-DNA-peptide complex by flow cytometry. Fluorophores are either directly conjugated to the oligo at its 5′-end using a NHS Ester linker or a spacer of varying lengths. Attachment may occur, for example, by covalent linkage of puromycin to the 3′-end of the short oligo linker via multiple hexa-ethylene glycol spacer. It is thought that increasing the length of the spacer is most effective when multi-pass membrane proteins (GPCRs and ion-channels) are the targets or when the peptides are binding to a cavity or groove of a protein. For example, binding of an mRNA-DNA-peptide to a multi-span cell expressed protein (herein referred to as “Protein III”) complex, expressed on the cell surface of HEK cells, is better detected by flow cytometry as the spacer length is increased (Table 3). This might be due to enhanced accessibility of the fluorophore with the increased linker length. However, the data in Table 3 also demonstrate that a spacer might not be required for detection by flow cytometry.

TABLE-US-00003 TABLE 3 Detection of binding of mRNA-DNA-Protein III complex using modified puromycin linker. % MFI Receptor- Spacer HEK HEK* No spacer (direct conjugation of fluorophore to oligo) 73% 5% +C9 94% 4% Triethylene glycol spacer of MW of 212 D +C18 96% 3% 18-atom hexa-ethyleneglycol spacer of MW of 344.3 Da *HEK cells without exogenous receptor is used as a negative control.