Scharenberg Lab
Post on 21-Mar-2016
32 Views
Preview:
DESCRIPTION
Transcript
Scharenberg Lab• Technical update on yeast display (Hoku/Jordan)
– Overhang length can influence binding (see meeting notes)
– Half sites can be interrogated independently using yeast display (Hoku)
– Cleavage assay - optimized (Hoku)
• SCID target progress (Jordan)– specificity analysis of best designed variant– Results from 4 rounds of iterative mutagenesis/sorting
of best designed variant
Flow cytometric analysis of half site interactions
- base pair alterations in the I-Ani minus half site produce large affinity changes; whereas base pair alterations in the plus half site typically do not- Obvious explanation: the above is due to differences in the individual half site affinities- From an engineering standpoint, this facet of I-Ani makes masks our ability to do to detailed evaluation of plus half site designs and evolution
Strategy: - interrogate surface displayed I-Ani affinity with oligos which extend from central four across one half site
--> Can half sites be separately/individually interrogated?
Halfsite Dilutions
Concentration
Plus
Minus
MFI
Minus
Plus
Concentration (nM) Concentration (nM)
Concentration (nM) Concentration (nM)
MFI
MFI
MFI
MFI
WT Y2
E148D L156R
Plus site Kd = 12,383 nM
Minus site Kd = 503 nM
Plus site Kd = 4,591 nM
Minus site Kd = 128 nM
Plus site Kd = 4,071 nM
Minus site Kd = 116 nM
Plus site Kd = 9,539 nM
Minus site Kd = 373 nM
Thoughts and speculation:• I-Ani sits down on minus half site DNA, and the
plus half site flops in the breeze - if a compatible plus half site sequence of DNA is present, it cleaves? – Could this be why Ani structures are blurry?– Are we really separately interrogating the plus half
site (e.g. could minus half site-induced conformational changes be important?)
• What does Rosetta say about the plus half site? – Can we design a tighter plus half site binder which
would still cleave and give better structures?
WT Y2 E148D L156R
Overlay of Mg Ca
Flow cleavage assay optimization: (Hoku)
Min
us
Teth
ered
to
c-M
yc
Min
us
Unt
ethe
red
PE-SA (on minus half site)
Ale
xa-6
47 (o
n pl
us h
alf s
ite e
nd)
- Many failures with yeast for unclear reasons: (Jordan and Andy)- Half site data: Hoku noted we have been labeling the minus half which doesn’t come off easily, perhaps we would do better if we labeled the plus half site?
T=0
30 minutes post-incubation (at 37°C)
WT Y2 E148D (inactive) L156R
Overlay of Mg Ca
Conclusion: Hoku got it right, the minus half site sticks very well even after cleavage, while the plus dissociates rapidly after cleavage
PE-SA (on minus half site)
Ale
xa-6
47 (o
n pl
us h
alf s
ite e
nd)
Min
us
Teth
ered
to
c-M
yc
Min
us
Unt
ethe
red
30 min Ca tethered
0
20
40
60
80
100
120
0 20 40 60 80 100
WTE148DNon-cleaving
30 min Mg tethered
0
20
40
60
80
100
120
0 20 40 60 80 100
WTE148DNon-cleaving
30 min Ca untethered
0
20
40
60
80
100
120
0 20 40 60 80 100
WTE148DNon-cleaving
30 min Mg untethered
0
20
40
60
80
100
120
0 20 40 60 80 100
WTE148DNon-cleaving
WT / E148D mixing assays (post-incubation)
0 overhangs
Kd = 21,280 nM
1 overhang
Kd = 5,688 nM
2 overhangs
Kd = 2,301 nM
3 overhangs
Kd = 5,299 nM
4 overhangs
Kd = 1,297 nM
5 overhangs
Kd = 703 nM
Overhang WT binding (full WT Ani target plus variable length overhangs)
Gate
d on
Myc
+
Original Library on Y2 L156R background: sorted for Myc+ only
After 3 rounds sorting for dsAni SCID binding, epPCR/reshuffling library, unsorted Counterselection …
dsAn
i SCI
D-64
7
dsAni WT-PE
2.0.B1 2.1.C0 2.1.C2 2.1.C3 2.1.C4 2.1.C5
SCID Target
C0 C1 C2 C3 C4 C5
WT Target
C0 C1 C2 C3 C4 C5B1 B1
ds
ss
cut
SCID target design project
• Start with 160 designs (JH), shuffle in yeast with additional diversity from ep-PCR
Output: - no primary design came through selection- final population was dominated by a few designed STS motifs- no cleavage activity
TGAGGAGGTTTCTCTGTAAAAGGAAGGATTCTCTGTAA
WT AniSCID
Comparison of selection methods: direct vs. counterselection
Blue: SCID Oligo bindingRed: WT Oligo bindingBlack: ratio SCID/WTGreen: ratio WT/SCID
“You get what you select for”
SCID target: specificity analysis• We have binding selectivity for what we selected for and against what we selected against
(dsSCID vs. dsAni– what does that mean for selectivity against “one offs”?
SCID oligo: Blue: others are “one offs”
-10 -9 -8 -7
-6 -5 -4 -3
Conclusion: selectivity of best binder is dominated by -10 position “You get what you select for”
0
100
200
300
400
500
600
700
Start Round 2 Round 3
Median Fluorescence Intensity (normalized)
SCIDWT
Start Round 2 Round 3Mutagenized (no selection)
- moderately aggressive gating to avoid bottlenecking our evolutionary pool- mutagenesis every round- Result: moderate population level improvement in MFI
Improvement by evolution?
Future Directions
- refined selectivity via counterselection against one-offs (in progress)- redesign native plus half site?
D16E/A/G/V/N (38%)
L36R (19%)
S37G/N (19%)
K39E/R/N (12%)
I64T (enriched from 33% to 98%)
K60T/M (10%)
D73N (5%)
I53V/T (19%)
I44V (depleted from 33% to 5%)
mutually exclusive
Round 4 output
Red areas of DNA - altered base pairs
top related