1 of 53

BENG 247A

Supreresolution and multiplexed imaging Bogdan Bintu

2 of 53

Molecules within a human cell

Dense organization within a cell

3x10⁹bases

~150,000

regulatory regions

30,000 genes

Need of 10,000-100,000 colors

Need of nm scale resolution

3 of 53

Fourier optics

Point

source

2-slit screen

camera

Photons

(rays)

4 of 53

Basic Microscope

Camera plane

Sample plane

Objective

Optical axis

Tube lens

Camera

5 of 53

The Abbe Diffraction Limit

The Abbe diffraction limit found in 1873 by Ernst Abbe.

6 of 53

Fluorescence microscopy

Image within a cell

~30,000 types of proteins

7 of 53

Fluorescence microscopy

Image within a cell

8 of 53

Super-resolution fluorescence microscopy

Periodic structure of the axonal skeleton

Xiaowei Zhuang

PALM

STORM

9 of 53

Super-resolution fluorescence microscopy

Bleach the sample (Dark state) +Glucose oxidase +BME

Fluorescence imaging

Molecular positions

Cy5

Energy Levels

Ψ₀

E₀

Ψ₁

E₁

Ψ₂

E₂

ground state

10 of 53

Super-resolution fluorescence microscopy

Epifluorescence imaging of DNA

Nucleus

Cy5 and Alexa405

Bleach the sample (Dark state) +Glucose oxidase +BME

11 of 53

Super-resolution fluorescence microscopy

12 of 53

Super-resolution fluorescence microscopy

13 of 53

Super-resolution fluorescence microscopy

14 of 53

Super-resolution fluorescence microscopy

Active Inactive Repressed

Nature, 2016, Super-resolution imaging reveals distinct chromatin folding for different epigenetic states

Alistair N. Boettiger, Bogdan Bintu, Jeffrey R. Moffitt, Siyuan Wang, Brian J. Beliveau, Geoffrey Fudenberg, Maxim Imakaev, Leonid A. Mirny, Chao-ting Wu & Xiaowei Zhuang

15 of 53

Super-resolution fluorescence microscopy

Cell, 2015, Mapping Synaptic Input Fields of Neurons with Super-Resolution Imaging

Yaron M. Sigal, Colenso M. Speer,Hazen P. Babcock and Xiaowei Zhuang

16 of 53

Super-resolution fluorescence microscopy - live imaging

Elife, 2014, Developmental mechanism of the periodic membrane skeleton in axons

Guisheng Zhong, Jiang He, Ruobo Zhou, Damaris Lorenzo, Hazen P Babcock, Vann Bennett, Xiaowei Zhuang

17 of 53

Next generation sequencing

4¹⁰⁰>10⁶⁰ > Number of atoms in the Sun

3x10⁹bases

~150,000

regulatory regions

30,000 genes

Need of 10,000-100,000 colors

18 of 53

Microfluidics

Microscope lasers

Spatial transcriptomics (Method of the year in 2020 - Nature methods)

sample

5 fluorescent colors > 1,000-10,000 effective colors

Multiplex Error Robust In Situ Hybridization (MERFISH)

Chen et al Science 2015

X. Zhuang, Nature Methods, 2021

Advances in DNA synthesis (labelling thousands of molecular types)

Combinatorial encoding and decoding scheme

+

=

Next generation sequencing

19 of 53

Microfluidics

Microscope

Spatial transcriptomics/genomics

MERFISH (Zhuang lab)

seqFISH (Cai lab)

Spatial transcriptomics (Frisén lab)

SlideSeq (Chen&Makasco labs)

FISSEQ (Church lab)

STAR-map (Deisseroth lab)

High sensitivity (>80% efficiency) and spatial resolution (subcellular)
Low genomic throughput

(~500 colors)

Low sensitivity and spatial resolution

High genomic throughput (>30,000 colors)

Ståhl et al. Science 2016

Rodriques et al Science 2019

Chen et al Science 2015

Lubeck et al., 2014

Wang et al Science 2019

Lee et al Science 2014

Next gen. sequencing

Chen et al Science 2015 X. Zhuang, Nature Methods, 2021

20 of 53

RNA/DNA labelling

ssDNA/RNA

RNA

T

C

G

A

G

C

>15 bases - stable in PBS

In Situ Hybridization ( ISH)

Fluorescence F

21 of 53

RNA/DNA labelling

Rob Singer

…

single-molecule FISH

20bases

>30 probes

22 of 53

DNA synthesis

https://www.youtube.com/watch?v=KUm173PZJBQ

voltage/micromirrors redirecting photons

Change in Ph - incorporation of a nucleotide

23 of 53

RNA/DNA labelling

…

single-molecule FISH

24 of 53

Multiplexed RNA/DNA labelling

…

single-molecule FISH

Readout sequences

Readout probe 1

2

1

mRNA1

25 of 53

Multiplexed RNA/DNA labelling

…

single-molecule FISH

Readout sequences

Readout probe 2

…

2

1

mRNA1

mRNA2

26 of 53

Encoding probes

Readout 3

...

Multiplexed RNA imaging - sequential labelling

N readouts

N species of RNA

Readout

sequence

Targeting

sequence

Readout 1

Readout 2

Add smFIsh reference from rob singer as Xiaowei.

The goal is the following: how can we efficient image and distinguish many different species of RNA expressed within the same cells?

We start by hybridizing to the RNA small probes that have two components: a targeting region complementary to the RNA of interest and hanging sequences called readout sequences which will allow the rapid hybridization of fluorescent probes.

One strategy is to design unique readouts sequences for each RNA type, such that by flowing a fluorescent readout probe we labelel and image RNA1. Then we extinguish this signal and label RNA2 and so on. We sometimes employ this method when the number of genes is small. But as the number of genes of interest increases to 1,000, as in the case of olfactory receptors, it becomes impractical to flow 1000 readout probes one-by-one and a more efficient method is needed.

27 of 53

Readout 2

Readout 3

...

Readout 1

1

0

10

11

01

101...

110...

011...

N = 16

>60,000

RNA species

RNA 1: 1 0 1 0 0 0 ... 0

RNA 2: 0 1 1 0 0 0 ... 0

RNA m: 1 1 1 1 1 1 ... 1

…

Readout: 1 2 3 4 5 6 ... N

N readouts ~ 2^N species of RNA

Multiplexed RNA imaging - combinatorial labelling

One alternative would be to use a combinatorial labelling scheme.

In this scheme we associate each of the RNA species with a binary barcode across all the readouts probes.

And then we readout out this barcode sequentially for each RNA molecule as we flow the different readout probes.

For instance for RNA 1 which has barcode 101 we will measure a 1 when flowing readout probe 1 because this fluorescent probe bound to it.

To illustrate improvement, I will mark with a 1 if the RNA molecule is fluorescent in that round and 0 if it is not.

Then across readout rounds RNA 1 will have a unique binary code: 1(because it fluoresces when adding readout 1)0 (because it does not fluoresce in this round) 1 and so on.

We can then make a table with all the possible codes which we can assign to RNAs using strategy which is effectively all the binary codes of length N (readout rounds).

This is great because then the encoding power (the number of RNA species grows exponentially with the number of rounds.) So that with only 16 rounds of readout labelling an imaging we can distinguish >60,000 RNA species, the whole transcriptome.

28 of 53

RNA 100: 1 0 1 1 1 0 0 0 0 0 0

RNA 200: 1 0 0 1 1 0 0 0 0 0 0

N = 16

>50%

RNA is called incorrectly

Readout 2

Readout 3

...

Readout 1

1

0

10

11

01

101...

110...

011...

RNA 1: 1 0 1 0 0 0 ... 0

RNA 2: 0 1 1 0 0 0 ... 0

RNA m: 1 1 1 1 1 1 ... 1

…

Readout: 1 2 3 4 5 6 ... N

Multiplexed RNA imaging - combinatorial labelling

29 of 53

Readout 2

Readout 3

...

Readout 1

1

0

10

11

01

101...

110...

011...

RNA 1: 1 0 1 0 0 0 ... 0

RNA 2: 0 1 1 0 0 0 ... 0

RNA m: 1 1 1 1 1 1 ... 1

…

Readout: 1 2 3 4 5 6 ... N

RNA 100: 1 0 1 1 1 0 0 0 0 0 0

RNA 200: 1 0 1 0 0 0 0 1 1 0 0

1 0 0 1 1 0 0 0 0 0 0

MERFISH:

Multiplexed, error-robust Fluorescence in situ Hybridization

Chen et al, Science (2015)

Multiplexed RNA imaging - combinatorial labelling

30 of 53

Multiplexed RNA/DNA labelling

Chen et al, Science (2015)

MERFISH images across readouts

~120 genes were targeted in 16 bits

31 of 53

MERFISH - Bintu lab applications

Imaging chromatin regulation at the genome scale

Su,...,Bintu*,Zhuang* Cell 2020

Inducing neurogenesis

Collaboration with Don Cleveland

The regulation of transcription factors across development

Collaboration with Chris Glass

Olfactory epithelium(nose)

Olfactory bulb (brain)

Receptor

RNA

Epithelium

Glomeruli

The connectome of the olfactory system

Bintu et al 2023 (in preparation)

32 of 53

dying

convert

Inducing neurogenesis

In the adult centers, the neural paths are something fixed and immutable:

everything may die, nothing may be regenerated

Ramon y Cajal, 1928

*Exception: Adult neurogenesis

Gage lab, Doetsch lab, Linnarsson lab and more

33 of 53

Approaches to glia-to-neuron transdifferentiation

A therapeutically viable approach

34 of 53

What cell types convert into neurons?

Which types of neuron can be obtained and in what brain regions?

Can we capture intermediate states?

Maimon et al, Nat Neurosci, 2021

+Tamoxifen

Glia promoter

Genetically label glial cells

New neurons after 2 months

+PTB-ASO

PTB mRNA

Glial cells

Improved memory

35 of 53

BIT 1

BIT 2

...

Thalamus

Dentate

gyrus

MERFISH in the brain

BIT 22

36 of 53

Dentate

gyrus

Thalamus

BIT 1

BIT 2

...

BIT 22

BIT 1

BIT 2

BIT 3

BIT 6

BIT 9

BIT 20

...

BSN (Bassoon Presynaptic Cytomatrix Protein)

1

0

1

MERFISH in the brain

37 of 53

38 of 53

Astrocytes

Cortical excitatory neurons

Cortical inhibitory neurons

CA2

CA1

Oligodendrocytes

DG

Pericytes

CA3

Endothelial

Thalamic neurons

MERFISH (230 genes)

Single-cell UMAP (~57,000 cells)

39 of 53

Ependymal cells

Immature SVZ neurons

Mature Dentate gyrus neurons

Astrocytes

Immature dentate gyrus neurons

Zhao, Deng, Gage, Cell, 2008

SGZ

40 of 53

Control (CBM4)

SOX11

Number of transcripts per cell

0

>20

Zhao, Deng, Gage, Cell, 2008

SGZ

41 of 53

Control (CBM4)

IGFBPL1

(Growth factor modulator)

Number of transcripts per cell

0

>20

Zhao, Deng, Gage, Cell, 2008

SGZ

42 of 53

Mki67

(cell-cycle –G2/M)

Number of transcripts per cell

0

Zhao, Deng, Gage, Cell, 2008

SGZ

43 of 53

Progenitors

Immature neurons

Dentate gyrus neurons

Astrocytes

8 week mouse

Immature neurons

1 year mouse

Neurogenesis stops in old mice

44 of 53

Progenitors

Immature neurons

Dentate gyrus neurons

Astrocytes

8 week mouse

Immature

neurons

1 year mouse + PTBP-ASO (2weeks)

Inducing neurogenesis in old mice via PTB-ASO

45 of 53

3 days post PTBP1-ASO injection

1 week post PTBP1-ASO injection

1 Month post PTBP1-ASO injection

Dentate gyrus

Subventricular zone

Igfbpl1 – Insulin like growth factor binding protein 1

Dentate gyrus

Subventricular zone

Dentate gyrus

Subventricular zone

Timepoints for after PTB-ASO delivery

46 of 53

Progenitors

Immature neurons

Dentate gyrus neurons

Astrocytes

8 week mouse

Immature

neurons

1 year mouse + PTBP-ASO (2weeks)

Inducing neurogenesis in old mice via PTB-ASO

Immature neurons

1 year mouse

47 of 53

Conclusions

PTBP1 suppression via ASO reactivates the neurogenesis pathway in an otherwise dormant subependymal (Stage 1)

Some of these cells upon cell division incorporated into the striatum as inhibitory mature neurons

Human SVZ contains these dormant PTBP1 positive progenitor cells

Stage 1

Stage 2

Stage 3

Stage 1

Stage 2

Stage 3

1 year old mice

Saline injection

PTBP1-ASOs

Next direction

Application to Huntington’s Disease

Direct astrocyte glia-to-neuron conversion

48 of 53

Funding sources and collaborators

Don Cleveland

Roy Maimon

Carlos Marinas

Quan Zhu

49 of 53

50 of 53

Subventricular zone

Where PTBP1 is most highly expressed in the mouse brain?

Number of transcripts per cell

0

51 of 53

Where PTBP1 is most highly expressed in the mouse brain?

Number of transcripts per cell

0

Astrocytes

Choroid

Plexus

Ependymal

Ventricular

Ependymal

Oligos

Cortical

Neurons

Striatal

Neurons

8 weeks old mouse

Immature

Neurons

Dentate

Gyrus

0

11>

20>

52 of 53

Astrocytes

Choroid

Plexus

Ependymal

Ventricular

Ependymal

Oligos

Cortical

Neurons

Striatal

Neurons

Astrocytes

Choroid

Plexus

Ependymal

Ventricular

Ependymal

Oligos

Cortical

Neurons

Striatal

Neurons

1 year old mouse 2 weeks post PTBP1-ASO

1 year old mouse control saline

Dentate

Gyrus

Dentate

Gyrus

Immature

Neurons

Immature

Neurons

0

11>

Highest level of suppression in the Ependymal cells

53 of 53

Using Aldhl1:Sun1GFP mouse model marks Astrocytes and Ependymal cells

Sun1 – nuclear envelope protein

DAPI

Dentate gyrus

Sun1GFP

DAPI

Sun1GFP

Choroid Plexus

Choroid plexus

Composed of Ependymal cells