Protocol

Gateway-Compatible Yeast One-Hybrid Screens

Program in Gene Function and Expression and Program in Molecular Medicine, University of Massachusetts Medical School, Worcester, MA 01605, USA

¹Corresponding author (marian.walhout{at}umassmed.edu)

INTRODUCTION

Protein-DNA interactions (PDIs) between transcription factors (TFs) and their target genes form the backbone of transcription regulatory networks. Such PDIs can be identified with either a TF or a gene as a starting point. The Gateway-compatible yeast one-hybrid (Y1H) system provides a high-throughput, gene-centered method for the identification of interactions between a "DNA bait" (e.g., cis-regulatory DNA elements or gene promoters) and "protein preys" (e.g., TFs). The Y1H system is a genetic system to detect PDIs based on selection of reporter gene expression in yeast. DNA baits are fused by Gateway cloning to two reporter genes, HIS3 and lacZ, and the resulting DNA bait::reporter constructs are subsequently integrated into the genome of the host yeast strain. After integration, baits are examined for self-activation (i.e., their ability to drive reporter gene expression in the absence of an exogenous prey protein). Subsequently, each DNA bait is screened for interacting proteins by transforming a library of preys into the corresponding Y1H DNA bait yeast strain. Preys are hybrid proteins composed of a protein from the organism of interest and a heterologous transcription activation domain. When a prey protein binds to the DNA bait, the heterologous activation domain activates reporter gene expression. Thus, physical interactions between both repressors and activators and their DNA targets can be identified.

MATERIALS

Reagents

Agar (US Biologicals)

Amino acid mix

Chloroform (Sigma)

DB3.1 bacteria (Invitrogen)

dNTPs (Invitrogen)

Gateway BP clonase II enzyme mix (Invitrogen)

Gateway LR clonase II enzyme mix (Invitrogen)

Glycerol (EMD)

Liquid nitrogen (LN₂)

ß-Mercaptoethanol (Sigma)

Phenol-chloroform (EMD)

Platinum Taq DNA Polymerase High Fidelity (Invitrogen)

Polyethylene glycol 3350 (PEG; Fisher)

Primers:

M13FW, 5'-GTAAAACGACGGCCAGT-3'

1HIFW, 5'-GTTCGGAGATTACCGAATCAA-3'

AD, 5'-CGCGTTTGGAATCACTACAGGG-3'

HIS293RV, 5'-GGGACCACCCTTTAAAGAGA-3'

LacZ592RV, 5'-ATGCGCTCAGGTCAAATTCAGA-3'

TERM, 5'-GGAGACTTGACCAAACCTCTGGCG-3'

FW DNA bait-specific PCR primer (see Step 24)

Restriction enzyme AflII or XhoI (see Step 10)

Restriction enzyme NcoI or ApaI (see Step 10)

Restriction enzymes BglII and SalI (see Step 84)

Salmon sperm DNA (10 mg/ml; US Biologicals)

Selective yeast medium (see Steps 50-56)

Taq DNA polymerase (NEB)

T4 DNA ligase (NEB)

TBE solution

TE (pH 8.0)

TE/lithium acetate solution

TE/lithium acetate/PEG solution

X-Gal solution

Yeast strain YM4271 (MATa, ura3-52, his3-200, ade2-101, ade5, lys2-801, leu2-3, 112, trp1-901, tyr1-501, gal4D, gal8D, ade5::hisG) (Clontech)

YEPD medium, liquid (2 liters) with 1 g adenine sulfate added

YEPD medium plus 2% Bacto agar (2 liters)

Pour into 15-cm sterile Petri dishes (~80 ml/dish). Dry for 3-5 days at room temperature, wrap in plastic bags or foil, and store at room temperature.

Z-buffer

Zymolyase lysis buffer

Equipment

96-well PCR plates (Fisher)

Cone buckets (250 ml; Corning)

Glass beads (Fisher)

Nitrocellulose filters (GE Osmonics)

Petri dishes (15 cm; VWR)

Replica-plating block

Velveteen squares (see Step 49)

Water bath, preset to 42°C (see Step 43)

Whatman paper (Fisher)

The reagents and methods in this protocol were developed for Caenorhabditis elegans but can be adapted for use in other systems.

METHOD

An outline of the entire Y1H procedure is shown in Figure 1 .

View larger version (13K): [in this window] [in a new window]

Figure 1. Schematic overview of Y1H assays

Making Y1H DNA Baits

For Gateway cloning of DNA baits into our Y1H Destination vectors, baits need to be available as Entry clones. DNA bait Entry clones can be obtained from promoterome resources (Dupuy et al. 2004) or created ab initio (Deplancke et al. 2006).

Cloning DNA Baits into Y1H Destination Vectors (3-4 days)

Previously, we used Multisite Gateway cloning for the creation of DNA bait::reporter fusion constructs (Deplancke et al. 2004). However, we noticed that the efficiency of such reactions is relatively low compared to traditional Gateway reactions (data not shown). To circumvent efficiency issues, we created novel Y1H Destination vectors that can be used in conventional Gateway LR cloning reactions (http://www.invitrogen.com/content/sfs/manuals/lr_clonase_man.pdf; see Fig. 1). These vectors, pMW#2 and pMW#3, contain a Gateway cassette with AttR4 and AttL1 recombination sites and either a HIS3 (pMW#2) (see Fig.2A ) or a lacZ (pMW#3) (see Fig. 2B) reporter gene.

View larger version (11K): [in this window] [in a new window]

Figure 2. Novel Y1H destination vectors. (A) pMW#2; (B) pMW#3

1. Take the Gateway LR clonase II enzyme mix from the -80°C freezer and place it on ice. Compose all reactions on ice.

2. Combine in sterile tubes: ~200 ng of pMW#2 to generate DNA bait::HIS3 constructs or ~200 ng of pMW#3 to generate DNA bait::lacZ constructs (should be 1 µl), 1 µl of Gateway LR clonase II enzyme mix, and enough DNA bait Entry clone miniprep (e.g., from promoterome resources) to obtain a final volume of 5 µl. As a negative control, prepare an identical LR mix without Entry clone but with TE buffer (pH 8.0) instead.

3. Incubate overnight at 25°C.

4. Transform the entire reaction mix into 50 µl of Escherichia coli DH5 thermocompetent cells (>10⁷ transformants per microgram of DNA), plate onto LB-ampicillin plates (100 µg/ml), and incubate them overnight at 37°C. The number of colonies obtained usually varies between 10 and 200. The negative control should give rise to no or only a few colonies (less than five).

5. Verify the insert size of the Destination clones by PCR of at least two colonies per construct using M13FW and HIS293RV primers for DNA bait::HIS3 constructs and the 1HIFW and LacZ592RV primers for DNA bait::lacZ constructs (see Fig. 2). With a sterile pipette tip, pick a single E. coli colony, patch the cells on a fresh LB-ampicillin plate for future use, and transfer the remaining cells to the appropriate PCR tube or well containing 25 µl of PCR mix consisting of 1 unit of Taq DNA polymerase, 0.5 µl of each 25 µM primer, 2.5 µl of 10X NEB PCR buffer, and 0.5 µl of 10 mM dNTP. The PCR program should be as follows:

i. 94°C, 2 minutes

ii. 94°C, 1 minute

iii. 56°C, 1 minute

iv. 68°C, 1 minute per kilobase of promoter sequence

v. Repeat from Substep 5.ii, 34 cycles

vi. 72°C, 7 minutes

6. Identify the colonies yielding a PCR product of the expected size by running 5-10 µl of PCR reaction on a 1% (w/v) agarose gel in 1X TBE. For HIS3 constructs, the size of PCR amplicons should be the length of the DNA bait plus ~400 bp. For lacZ constructs, the size of PCR amplicons should be the length of the DNA bait plus ~800 bp (see Fig. 2).

7. Incubate the LB-ampicillin plate overnight at 37°C.

8. The next day, pick a single colony containing the correct DNA bait Destination clone, and grow it in 2 ml of liquid LB-ampicillin.

9. Purify the Destination clone DNA by miniprep for subsequent integration into the yeast genome.

Integrating the DNA Bait Constructs into the Y1H Yeast Strain (6-7 days)

DNA bait::HIS3 and DNA bait::lacZ constructs are sequentially integrated into the mutant HIS3 and URA3 loci of YM4271. To facilitate homologous recombination, both DNA bait constructs first need to be linearized.

10. Digest 15 µl of miniprep (~1 µg of DNA bait::HIS3 constructs) with either AflII or XhoI, and a similar amount of DNA bait::lacZ constructs with either NcoI or ApaI in a 20-µl reaction volume. Make sure that the restriction enzyme of choice does not cut within the DNA bait sequence.

11. Verify linearization of constructs by running 1-2 µl of the restriction digest reaction mixture on a 1% (w/v) agarose gel in 1X TBE next to an equal amount of the undigested constructs.

12. Transform linearized DNA bait::HIS3 constructs into YM4271 using the high-efficiency transformation protocol that is also used for Y1H screens (see "High-Efficiency Yeast Transformations," Steps 27-46, below). Include a "no DNA" negative control. Plate transformation reactions on Sc-His medium.

13. Incubate for 3 days at 30°C.

14. Harvest all colonies for the next integration step, and grow this mixed pool on Sc-His plates at 30°C. The negative control should give rise to no colonies.

15. Transform the linearized DNA bait::lacZ construct into the corresponding DNA bait::HIS3-containing yeast strain (from Step 14) using the high-efficiency transformation protocol, and plate on Sc-His,-Ura medium. Include a "no DNA" negative control.

16. Incubate for at least 3 days at 30°C. The number of colonies obtained usually varies between 50 and 1000. The negative control should give rise to no colonies.

Verifying Integration of the DNA Baits into Yeast by PCR (1 day)

We found that yeast colony PCR from YM4271-based strains is considerably more difficult than we previously experienced with the yeast two-hybrid strains MaV103 and MaV203 (Walhout and Vidal 2001). To solve this, we optimized a PCR protocol that is based on lysis of the yeast cells with Zymolyase.

17. Grow yeast on solid YEPD media overnight at 30°C. We find that growing on YEPD, rather than on selective medium, increases the PCR efficiency.

18: Aliquot 15 µl of Zymolyase enzyme suspension into the wells of a 96-well plate. Since the enzyme has low solubility, it is important to mix the suspension thoroughly before distributing it into the wells.

19. Using sterile toothpicks or pipette tips, gently remove approximately one-fourth of a match head of the yeast colonies, and transfer it to the wells.

20. Incubate the yeast-enzyme mix for 30 minutes at 37°C.

21. Heat-inactivate the enzyme for 10 minutes at 95°C (in a PCR machine).

22. Pellet cell debris by centrifugation at 700g for 5 minutes.

23. Dilute the lysate by adding 85 µl of sterile H₂O. The lysate can be stored at -20°C for subsequent PCR reactions.

24. Set up a 50-µl PCR reaction containing 1 µl of 10 mM dNTP, 5 µl of 10X Platinum Taq polymerase PCR reaction buffer, 1 µl of FW DNA bait-specific PCR primer (25 µM), 1 µl of HIS293RV (25 µM for DNA bait::HIS3 constructs) or LacZ592RV (25 µM for DNA bait::lacZ constructs), 1 unit of Platinum Taq polymerase, and 5 µl of yeast lysate.

25. The PCR program should be as follows:

i. 94°C, 3 minutes

ii. 94°C, 1 minute

iii. 56°C, 1.5 minutes

iv. 68°C, 3.5 minutes

v. Repeat from Substep 25.ii, 34 cycles

vi. 68°C, 5 minutes

26. Run 10 µl of the PCR reactions on a 1% (w/v) agarose gel in 1X TBE. For HIS3 constructs, the size of PCR amplicons should be the length of the DNA bait plus ~300 bp. For lacZ constructs, the size of PCR amplicons should be the length of the DNA bait plus ~700 bp.

High-Efficiency Yeast Transformations (1 day)

27. Patch the relevant yeast strain on an appropriate yeast medium plate (YEPD for YM4271 or Sc-His for DNA bait::lacZ integrations).

28. Incubate overnight at 30°C.

29. Resuspend the strain(s) in 50 ml of liquid YEPD to obtain a starting culture with an OD₆₀₀ of ~0.15-0.20.

30. Incubate for ~5 hours at 30°C, at 200 rpm in a shaking incubator until the culture reaches an OD₆₀₀ of 0.4-0.6.

31. Harvest the cells by centrifugation (700g) at room temperature for 5 minutes.

32. Meanwhile, boil a tube of salmon sperm DNA (10 mg/ml) for 5-10 minutes, and put it on ice.

33. From Step 31, decant the medium, and wash the cells in a volume of sterilized H₂O corresponding to 10% of the original YEPD volume.

34. Shake the bucket gently to resuspend the cells.

35. Centrifuge (700g) at room temperature for 5 minutes, and discard the supernatant.

36. Gently resuspend the cells in a volume of TE/lithium acetate solution corresponding to 2% of the original culture volume, centrifuge as above, and discard the supernatant.

37. Resuspend the cells in a volume of TE/lithium acetate solution corresponding to the volume of the original culture times the final OD₆₀₀ value divided by 100. (For example, if the final culture was 50 ml and had a final OD₆₀₀ of 0.5, resuspend in 0.25 ml of TE/lithium acetate solution.)

38. Add 1/10 volume of salmon sperm DNA (boiled; see Step 32, above).

39. For each transformation reaction, put 100 µl of cells in a 1.5-ml Eppendorf tube.

40. Add 18 µl of the linearized DNA bait::reporter construct to be transformed.

41. Add 600 µl of TE/lithium acetate/PEG solution, and resuspend carefully.

42. Incubate for at least 30 minutes at 30°C.

43. Heat-shock the cells for exactly 20 minutes at 42°C (in a water bath).

44. Centrifuge the tubes for 5 seconds in a microfuge at full speed at room temperature; remove the supernatant.

45. Resuspend the cells in 300 µl of sterile H₂O, and plate onto the appropriate selective media plates using glass beads.

46. Incubate at 30°C.

Testing the Self-Activation Properties of DNA Bait Yeast Strains

We generally observe two types of self-activation. First, reporter gene expression may be activated constitutively, for instance, when a DNA bait interacts with an endogenous yeast activator. Second, self-activation may not be constitutive (i.e., we observe it for some but not all integrants corresponding to an individual DNA bait). We observe this type of self-activation more frequently and found that it can occur when multiple copies of a DNA bait construct integrate into the yeast genome (Deplancke et al. 2004).

47. To circumvent high levels of self-activation, select a single integrant that exhibits the lowest levels of self-activation.

48. To test self-activation levels of each DNA bait, pick 12-24 colonies and test them for activation of both the HIS3 and the lacZ reporters.

Replica Plating

49. To transfer yeast colonies/spots from one plate to another, we use "velvets" made from velveteen to which yeast cells adhere, and a replica-plating block to transfer yeast from a plate onto a velvet. We cut the velveteen in 22 x 22 cm squares. Velvets are sterilized by autoclaving after each use, washed in a washer without soap, and dried in a dryer. After that, the velvets are stacked (~40/stack), wrapped in aluminum foil, and sterilized again.

Testing HIS3 and lacZ Reporter Self-Activation (~5 days)

Self-activation levels of the HIS3 reporter can be tested by growth on medium lacking histidine and containing 3AT, a competitive inhibitor of the His3p enzyme. The concentration of 3AT at which self-activation is minimal is then used for Y1H screening (see below).

50. Pick 12-24 individual colonies, and patch them in a 96-well spot format, as shown in Figure 3 , onto an Sc-His,-Ura plate; incubate them for 1-2 days at 30°C.

View larger version (32K): [in this window] [in a new window]

Figure 3. Overview of Y1H screens. (A) Self-activation test. (C) Controls; (1) p53BS; (2) Pmyo-2; (3) Pfog-3; (4) Phlh-8. (Left panel) Permissive plate (Sc-His,-Ura); (middle panel) selective plate (Sc-His,-Ura + 20 mM 3AT); (right panel) ß-Gal assay. White squares indicate a lowly self-activating colony that was chosen for subsequent Y1H experiments. (B) Y1H screen plate (Sc-His,-Ura,-Trp + 20 mM 3AT). White square indicates a positive colony. (C) Testing Y1H positive phenotypes. AD - Y1H bait transformed with an empty AD plasmid for comparison. (C) Controls; (1) p53BS with AD; (2) p53BS with AD-p53; (3-5) Pfog-3 with AD, AD-TRA-1, or AD-TRA-1DB. (Left panel) Permissive plate (Sc-His,-Ura,-Trp); (middle panel) selective plate (Sc-His,-Ura,-Trp + 20 mM 3AT); (right panel) ß-Gal assay. White squares indicate a double positive spot. (D) Yeast colony PCR. (E) Gap-repair. (C) Controls; (1) no DNA; (2) pPC86; (3) linear pPC86. (Left panel) Permissive plate (Sc-His,-Ura,-Trp); (middle panel) selective plate (Sc-His,-Ura,-Trp + 20 mM 3AT); (right panel) ß-Gal assay. White squares indicate a double positive spot.

51. To determine the extent of self-activation, we generated control DNA bait yeast strains that exhibit, respectively, low, medium, or high self-activation levels (see Fig. 3). Add these controls to the Sc-His,-Ura plate in Step 50 if available.

52. Replica-plate the spots onto the following plates: a fresh Sc-His,-Ura plate, Sc-His,-Ura + 3AT plates (containing 20, 40, 60, 80, and 100 mM 3AT). To perform ß-Gal assays, replica-plate the spots onto a nitrocellulose filter that has been placed on top of a YEPD plate. This ensures growth of the respective yeast colonies on the nitrocellulose filter (see "ß-Gal Assay," Steps 57-65, below.).

53. Replica-clean the 3AT-containing selective medium plates by replica-stamping onto a clean sterile velvet (which is not used further) until no yeast is visible (usually two times suffices).

54. Incubate the YEPD plate overnight at 30°C; the next day, perform a ß-Gal assay (see below).

55. Incubate the 3AT-containing plates for 3-10 days at 30°C. Monitor colony growth: Strong growth is an indication of self-activation.

56. DNA baits that constitutively grow on 100 mM 3AT cannot be used in Y1H assays. For the other baits, choose a colony that fails to grow (or grows minimally) at the lowest possible 3AT concentration and that is negative on the ß-Gal assay (see below).

ß-Gal Assay (1 day)

The ß-Gal assay is a colorimetric assay, based on blue-white coloring of the yeast: Yeast cells that express ß-Gal are blue, whereas yeast cells that do not express ß-Gal are white. Different gradations of blue indicate different levels of lacZ reporter gene expression. If possible, a colony that is white or very light blue should be picked for subsequent experiments.

57. Put two Whatman filters in an empty 15-cm Petri dish for each plate to be assayed.

58. Make a mix of 6 ml of Z-buffer, 11 µl of ß-mercaptoethanol, and 100 µl of 4% X-Gal per plate.
IMPORTANT: Make sure to do this in a hood.

59. Pour ~200 ml of liquid nitrogen into an ice bucket, cover the bucket, and place it in the hood.

60. Transfer 6 ml of the Z-buffer mixture onto each plate containing Whatman filters. Make sure the entire paper is soaked with buffer, and remove air bubbles using tweezers.

61. Take the nitrocellulose filter containing the yeast using the tweezers and place it in liquid nitrogen for 10 seconds.

62. Thaw the filter at room temperature by holding it in the air using tweezers.

63. Place the filter with the yeast facing up onto the Whatman filter, and remove air bubbles.

64. Incubate at 37°C.

65. Check for blue-white coloring regularly every hour during the first 4 hours, and take pictures. Continue the incubation overnight at 37°C, and check again for blue-white coloring the next day.

Y1H Prey Libraries

Different types of prey libraries can be used. We use libraries in which the prey is fused to the activation domain of Gal4p (AD). We routinely use an AD-cDNA library that is also commonly used for yeast two-hybrid screens (Walhout et al. 2000). In addition, we use an AD-TF mini-library that contains >60% of the 934 predicted C. elegans TFs (Deplancke et al. 2004; Reece-Hoyes et al. 2005). It should be feasible to make similar libraries for organisms of which many ORFs have been cloned and for which good sets of predicted TFs are available.

Screening Y1H DNA Baits (~10 days)

66. DNA bait strains are prepared as described above (see "High-Efficiency Yeast Transformations," Steps 27-46, above), except that the DNA bait strain is grown on Sc-His,-Ura plates overnight and then cultured in 500 ml of YEPD.

67. Transfer 2.5 ml of the TE/lithium acetate yeast suspension into a 15-ml tube, and add 1/10 volume of salmon sperm DNA (boiled, see above); mix well.
Also make an AD-only negative control (see Step 79, below): Take 100 µl of yeast suspension and add 3 ng of pPC86 (a "non-Gateway" AD-containing vector); add 500 µl of TE/lithium acetate/PEG solution, and mix well. Continue the transformation protocol from Step 72.

68. If screening an AD-TF library in addition to the AD-cDNA library, follow the steps listed below BEFORE adding the AD-cDNA library.

i. Take 500 µl of the yeast suspension from Step 67, and add 10 µg of AD-TF mini-library.

ii. Distribute the reaction mixture equally into five microcentrifuge tubes, add 500 µl of TE/lithium acetate/PEG solution, and mix well. Continue the transformation protocol from Step 72.

69. Add 30 µg of AD-cDNA library to 2.0 ml of yeast suspension, and mix well.

70. Add 12 ml of TE/lithium acetate/PEG solution, and mix by inverting the tube two or three times.

71. Distribute the mix equally into 15 microcentrifuge tubes.

72. Incubate the tubes for at least 30 minutes at 30°C.

73. Heat-shock for exactly 20 minutes at 42°C in an H₂O bath.

74. After heat shock, pool the AD-cDNA- and pool the AD-TF-containing transformation mixes separately (i.e., don’t mix AD-cDNA with AD-TF) into two 15-ml tubes, and centrifuge at 700g for 3 minutes. Remove the supernatant, and resuspend in, respectively, 5 ml (AD-cDNA) or 1.5 ml (AD-TF) of sterile H₂O.

75. Take 1/100 and 1/1000 dilution samples to determine the transformation efficiency, and plate out on Sc-His,-Ura,-Trp plates. We generally screen at least 1 x 10⁶ (cDNA library) or 3 x 10⁵ (AD-TF mini-library) yeast colonies.

76. Using sterile glass beads, plate the cells onto appropriate selective plates containing the 3AT concentration at which self-activation is minimal. Divide the transformation mixes over, respectively, 10 (AD-cDNA) and three (AD-TF) plates per Y1H screen.

77. Incubate at 30°C.

78. Pick colonies with a sterile toothpick after ~10 days, and transfer to Sc-His,-Ura,-Trp plates in a "spot" format (see Fig. 3).

79. To validate the strength of Y1H interaction phenotypes, several controls need to be added to the yeast plate generated in Step 78: (1) An AD-only negative control (i.e., pPC86) to compare the levels of reporter gene expression versus background reporter gene expression, (2) positive control colonies (if available) containing interactors that, respectively, induce low, medium, or high levels of reporter gene expression (see Fig. 3).
Please note that some DNA baits do not yield interactors when screened in the Y1H system (Deplancke et al. 2006). If no positives are retrieved, one could screen more colonies. Some Y1H screens yield hundreds of positive yeast colonies. This may reflect the high propensity of a bait to give rise to false-positive interactions, for instance, because it can be rather self-active. We advise that 96 positives be picked, retested by gap repair, and sequenced, although one could pick more positives as desired.

Identifying Double-Positive Yeast Colonies (3-4 days)

80. Replica-plate the positives (including the controls) onto the following plates: a fresh Sc-His,-Ura,-Trp plate, an Sc-His,-Ura,-Trp plate containing the 3AT concentration used for screening, and a YEPD plate containing a nitrocellulose filter for ß-Gal assays.

81. Proceed as described in "Testing HIS3 and lacZ Reporter Self-Activation" (Steps 50-56, above) and "ß-Gal Assay" (Steps 57-65, above).

82. Yeast colonies positive for both reporters are picked with a sterile toothpick and transferred to fresh Sc-His,-Ura,-Trp plates.

Yeast Colony PCR (1 day)

83. Yeast colony PCR should be performed using the PCR protocol to verify the DNA baits as described above ("Verifying Integration of the DNA Baits into Yeast by PCR," Steps 17-26), except that the general AD and TERM primers should be used.

Retesting Protein-DNA Interactions by Gap Repair (3-5 days)

To prevent the retrieval of false-positive interactions, all preys need to be retested by gap repair in fresh DNA bait-containing yeast cells.

This procedure can be done in a 96-well plate format.

84. Linearize pPC86 plasmid DNA (40 ng is used per gap repair reaction) using SalI and BglII (plasmid purification is not necessary).

85. Perform yeast colony PCR as described above (see "Verifying Integration of the DNA Baits into Yeast by PCR," Steps 17-26).

86. Patch the relevant Y1H bait onto a YEPD plate, and incubate overnight at 30°C.

87. Resuspend the yeast in 250-500 ml of liquid YEPD, depending on the number of gap repair reactions that need to be performed (400 ml of YEPD culture is needed for one 96-well plate), to an OD₆₀₀ of ~0.1, and proceed as in the yeast transformation protocol described above, except per reaction, transform 40 ng of linear pPC86 and 3-5 µl of yeast PCR product into 20 µl of competent yeast cells. Save the rest of the PCR reaction for sequencing (see below). Three negative controls without PCR product should be included: no DNA, 40 ng of cut, and 40 ng of uncut pPC86.

88. Plate reactions in a 96-well spot format onto Sc-His,-Ura,-Trp plates, and incubate at 30°C. The number of transformants should be an order of magnitude higher in gap-repaired samples compared to linear pPC86 alone. No transformants should be present in the no DNA control.

89. Replica-plate transformants onto Sc-His-Ura-Trp + 3AT and YEPD (plus nitrocellulose filter) plates, and incubate them at 30°C. Yeast colonies positive for both reporters are picked with a sterile toothpick and transferred to Sc-His,-Ura,-Trp plates. Negative controls should exhibit no or little growth.

Sequencing

90. PCR products corresponding to positive preys that retest by gap repair are sequenced using the AD primer to determine prey identity.

Scoring Y1H PDIs (1 day)

91. To circumvent the inclusion of false positives, we developed a stringent scoring system that is based on five different criteria (see Table 1, "A Y1H Scoring System"). A total score between 1 and 10 is obtained for each interaction by dividing the sum of the received points for all criteria by the highest number of points an interaction can get (i.e., 7) and multiplying it by a factor of 10. The higher the score, the higher the overall confidence in the PDI. By choosing a cutoff of 5, most false-positive interactions are eliminated.

Example 1: We take a DNA bait that exhibits high lacZ, but no HIS3 self-activation, for which one interactor (a known TF) was found from the AD-cDNA library.

Scoring: lacZ self-active (dark blue)? Yes 0 points HIS3 self-active (80 mM 3AT)? No 1 point Interactor-predicted TF? Yes 1 point Found multiple times? No 0 points Library scoring 1 AD-cDNA 1 point Total: 3 points Total score = (3/7 x 10) = 4.2 low confidence and thus likely a false positive.

Example 2: If the interactor would have been found three times with the same bait from the AD-TF library, then the scoring would have been as follows:

Scoring: lacZ self-active (dark blue)? Yes 0 points HIS3 self-active (80 mM 3AT)? No 1 point Interactor-predicted TF? Yes 1 point Found multiple times? Yes 2 points Library scoring 2 AD-TF 1 point Total: 5 points Total score = (5/7 x 10) = 7.1 high confidence.

Please note that highly self-active DNA baits tend to generate more false positives than low self-activators, since it is difficult to distinguish real positives from self-activating yeast colonies.

DISCUSSION

Gateway cloning facilitates the high-throughput generation of DNA baits using promoterome resources and Y1H reporter Destination vectors. TF-encoding open reading frames (ORFs) can be obtained from ORFeome resources, cloned into an AD-containing Destination vector (pDEST_AD), and combined to create an AD-TF mini-library. Because TF-encoding cDNAs are often under-represented in cDNA libraries and because the AD-TF library does not contain all predicted TF-encoding ORFs, DNA baits are best screened versus both a mini-library and a cDNA library. Since it is possible that false-positive PDIs may be obtained, it is critical to stringently filter the data to ensure the highest data quality.

The gene-centered Y1H system has several advantages over TF-centered methods such as chromatin immunoprecipitation, as it does not require anti-TF antibodies and as it allows the identification of PDIs involving TFs that are expressed at low levels or in a few cells in an organism. The Y1H system also allows the identification of novel putative TFs (i.e., proteins that have no predicted DNA binding domain) (Deplancke et al. 2006). Combining the Y1H system with Gateway cloning increased the throughput of the assay. Assuming that no technical problems arise, a single researcher should be able to complete the screening of 15-20 DNA baits from the initial cloning step to the final analysis of positives in 6-8 weeks using the protocols described here.

The Y1H system also has several limitations. First, several TFs need to be post-translationally modified to bind DNA or only bind DNA as heterodimers and are therefore missed in conventional Y1H screens (Deplancke et al. 2006). We recently generated a Gal4 AD-containing Gateway Destination vector containing a yeast selection marker complementary to that of pDEST_AD, which will allow the detection of DNA binding by obligate heterodimer complexes. Second, despite extensive filtering, some false-positive interactions may still be included in the final data set. A second assay can be used to independently validate Y1H interactions. For example, we have previously used transient cotransfections and luciferase reporter assays in mouse 3T3 cells to validate C. elegans Y1H protein-DNA interactions (Deplancke et al. 2004). Alternatively, a third Y1H reporter could be used, which has been applied successfully for Y2H screens (Walhout and Vidal 2001). Finally, the success of Y1H screens is highly dependent on the quality of the screening libraries. Taken together, the Gateway-compatible Y1H system provides a straightforward, high-throughput method to identify proteins that can bind to a DNA bait, or set of baits, of interest.

ACKNOWLEDGMENTS

We thank Inmaculada Barrasa, Christian Grove, and Job Dekker for discussions and Duffy Rasmussen for help preparing the manuscript. Research in the Walhout laboratory is supported by NIH grants (CA097516, DK068429, and DK071713) to A.J.M.W.

REFERENCES

Deplancke, B., Dupuy, D., Vidal, M., and Walhout, A.J.M. 2004. A Gateway-compatible yeast one-hybrid system. Genome Res. 14: 2093–2101.[Abstract/Free Full Text]

Deplancke, B., Mukhopadhyay, A., Ao, W., Elewa, A.M., Grove, C.A., Martinez, N.J., Sequerra, R., Doucette-Stamm, L., Reece-Hoyes, J.S., Hope, I.A., et al. 2006. A gene-centered C. elegans protein-DNA interaction network. Cell 125: 1193–1205.[Medline]

Dupuy, D., Li, Q.R., Deplancke, B., Boxem, M., Hao, T., Lamesch, P., Sequerra, R., Bosak, S., Doucette-Stamm, L., Hope, I.A., et al. 2004. A first version of the Caenorhabditis elegans promoterome. Genome Res. 14: 2169–2175.[Abstract/Free Full Text]

Reece-Hoyes, J.S., Deplancke, B., Shingles, J., Grove, C.A., Hope, I.A., and Walhout, A.J. 2005. A compendium of Caenorhabditis elegans regulatory transcription factors: a resource for mapping transcription regulatory networks. Genome Biol. 6: R110.[Medline]

Walhout, A.J.M. and Vidal, M. 2001. High-throughput yeast two-hybrid assays for large-scale protein interaction mapping. Methods 24: 297–306.[Medline]

Walhout, A.J.M., Sordella, R., Brasch, M.A., Temple, G., Hartley, J.L., Thierry-Mieg, N., and Vidal, M. 2000. Protein interaction mapping in C. elegans using proteins involved in vulval development. Science 287: 116–122.[Abstract/Free Full Text]