Journal of Physics D: Applied Physics
ACCEPTED MANUSCRIPT • OPEN ACCESS
Between life and death: strategies to reduce phototoxicity in super-
resolution microscopy
To cite this article before publication: Kalina L. Tosheva et al 2020 J. Phys. D: Appl. Phys. in press https://doi.org/10.1088/1361-6463/ab6b95
Manuscript version: Accepted Manuscript
Accepted Manuscript is “the version of the article accepted for publication including all changes made as a result of the peer review process,
and which may also include the addition to the article by IOP Publishing of a header, an article ID, a cover sheet and/or an ‘Accepted
Manuscript’ watermark, but excluding any other editing, typesetting or other changes made by IOP Publishing and/or its licensors”
This Accepted Manuscript is © 2020 IOP Publishing Ltd.
 
As the Version of Record of this article is going to be / has been published on a gold open access basis under a CC BY 3.0 licence, this Accepted
Manuscript is available for reuse under a CC BY 3.0 licence immediately.
Everyone is permitted to use all or part of the original content in this article, provided that they adhere to all the terms of the licence
https://creativecommons.org/licences/by/3.0
Although reasonable endeavours have been taken to obtain all necessary permissions from third parties to include their copyrighted content
within this article, their full citation and copyright line may not be present in this Accepted Manuscript version. Before using any content from this
article, please refer to the Version of Record on IOPscience once published for full citation and copyright details, as permissions may be required.
All third party content is fully copyright protected and is not published on a gold open access basis under a CC BY licence, unless that is
specifically stated in the figure caption in the Version of Record.
View the article online for updates and enhancements.
This content was downloaded from IP address 5.188.216.148 on 26/01/2020 at 06:47

----!@#$NewPage!@#$----
Between life and death: strategies to 
1 
reduce phototoxicity in Super-Resolution 
2 
Microscopy 
3 
 
4 
Kalina L. Tosheva1, Yue Yuan1, Pedro Matos Pereira2, Siân Culley1,3, and Ricardo Henriques1,3 
5 
1MRC Laboratory for Molecular Cell Biology, University College London, London, UK 
6 
2ITQB-NOVA, Oeiras, Portugal 
7 
3The Francis Crick Institute, London, UK 
8 
 
9 
Super-Resolution Microscopy enables non-invasive, molecule-specific imaging of the internal structure and 
10 
dynamics of cells with sub-diffraction limit spatial resolution. One of its major limitations is the requirement for high-
11 
intensity illumination, generating considerable cellular phototoxicity. This factor considerably limits the capacity for 
12 
live-cell observations, particularly for extended periods of time. Here, we give an overview of new developments in 
13 
hardware, software and probe chemistry aiming to reduce phototoxicity. Additionally, we discuss how the choice of 
14 
biological model and sample environment impacts the capacity for live-cell observations. 
15 
 
16 
Phototoxicity | Photodamage | Super-Resolution Microscopy | Fluorescence 
17 
Correspondence: s.culley@ucl.ac.uk, r.henriques@ucl.ac.uk  
 
18 
Page 1 of 28
AUTHOR SUBMITTED MANUSCRIPT - JPhysD-122201.R1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Accepted Manuscript

----!@#$NewPage!@#$----
Introduction 
19 
The spatial resolution of an imaging system is 
20 
defined as the capacity to distinguish closely 
21 
separated features; in light microscopy, this is 
22 
limited by diffraction to ~ 200-300 nm 
23 
. Consequently, microscopy approaches developed 
24 
to achieve resolutions beyond this limit are termed 
25 
‘Super-Resolution Microscopy’ (SRM) [1]. SRM 
26 
techniques that have recently gained popularity, 
27 
such as Photoactivated Localisation Microscopy 
28 
(PALM) [2], Stochastic Optical Reconstruction 
29 
Microscopy (STORM) [3], Structured Illumination 
30 
Microscopy (SIM) [4] and Stimulated Emission 
31 
Depletion (STED) Microscopy [5], have enabled 
32 
biological discoveries inaccessible to conventional 
33 
microscopy [6]–[9]. Alongside increased spatial 
34 
resolution, SRM retains many desirable features of 
35 
light microscopy techniques, including molecule-
36 
specific labelling and the potential for live-cell 
37 
imaging, unavailable 
to 
other 
high-resolution 
38 
techniques, such as electron microscopy. However, 
39 
the live-cell imaging potential of SRM has remained 
40 
largely untapped as the requirements of most SRM 
41 
techniques pose various challenges for exploring 
42 
dynamic processes under physiological conditions. 
43 
In contrast, such limitations are absent when using 
44 
fixed specimens. 
45 
Resolution increase in SRM is generally achieved at 
46 
the cost of high-intensity illumination [10]. These 
47 
requirements result in photobleaching, defined as 
48 
irreversible loss of fluorescence during imaging. 
49 
However, of greater importance to live-cell imaging 
50 
is sample health. Thereby, the dependency of SRM 
51 
on illumination intensities orders of magnitude 
52 
higher than conventional microscopy (W/cm2 - 
53 
GW/cm2 compared to mW/cm2 - W/cm2) makes 
54 
phototoxicity the biggest concern when employing 
55 
these techniques [10], [11]. In the context of 
56 
microscopy, 
phototoxicity 
is 
a 
broad 
term 
57 
encompassing physical and chemical reactions 
58 
caused by the interaction between light and cellular 
59 
components, with detrimental effects on the latter 
60 
[12], [13]. Correct biological interpretations from live-
61 
cell imaging can only be achieved if the observed 
62 
phenomena progress with minimal perturbation [14]. 
63 
A multitude of properties of the sample and the 
64 
imaging can influence phototoxicity and can thus be 
65 
optimised for improving SRM for live-cell imaging 
66 
(Fig. 1).  
67 
 
68 
Fig. 1 Summary of the factors that can be optimised to reduce 
69 
phototoxicity in Super-Resolution Microscopy. 
70 
On a molecular level, the main causes of 
71 
phototoxicity are photochemical processes that 
72 
directly damage intracellular components or lead to 
73 
the production of toxic products within the cell or in 
74 
its direct environment [15], [16]. The detrimental 
75 
effects of ultraviolet (UV) light on cells is particularly 
76 
well characterised; illumination with UV light can 
77 
trigger the so-called 'UV-response' (Fig. 2aFig. 1 
78 
Summary of the factors that can be optimised to reduce 
79 
phototoxicity in Super-Resolution Microscopy.) [17], [18], 
80 
Page 2 of 28
AUTHOR SUBMITTED MANUSCRIPT - JPhysD-122201.R1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Accepted Manuscript

----!@#$NewPage!@#$----
DNA-strand breaks [19], [20], and thymidine 
81 
dimerisations [21] (Fig. 2bFig. 1 Summary of the factors 
82 
that can be optimised to reduce phototoxicity in Super-
83 
Resolution Microscopy.), leading to mutations and 
84 
downstream apoptosis [22], [23]. Additionally, both 
85 
UV and visible wavelengths can excite other 
86 
endogenous photoactive molecules in the cell, such 
87 
as 
NAD(P)H 
[24], 
flavins 
[25], 
[26] 
and 
88 
porphyrins[27], [28]. Furthermore, in fluorescence 
89 
microscopy there are phototoxic effects associated 
90 
with the fluorescent molecules required for labelling 
91 
structures [15], [29]. Upon illumination, both 
92 
endogenous and exogenous photoactive molecules 
93 
can be excited to reactive states (most commonly 
94 
long-lived triplet states) capable of undergoing redox 
95 
reactions that lead to formation of reactive oxygen 
96 
species (ROS) (Fig. 2cFig. 1 Summary of the factors that 
97 
can be optimised to reduce phototoxicity in Super-Resolution 
98 
Microscopy.). ROS are considered the major 
99 
contributors to phototoxicity [12], [13]. Their 
100 
production can occur via direct reaction between the 
101 
excited molecule and environmental molecular 
102 
oxygen or via reactions with other neighbouring 
103 
molecules that generate free radicals [30]. ROS 
104 
have a broad range of negative effects ranging from 
105 
oxidising proteins, lipids, and DNA, as well as 
106 
systematic effects such as disrupting the redox 
107 
homeostasis, signalling pathways and cell cycle 
108 
[12], [31]. Notably, ROS production correlates with 
109 
illumination intensity and photobleaching [12], [15], 
110 
both of which are issues present in SRM. As a result, 
111 
there is considerable interest in developing SRM 
112 
technologies for improved sample health. Here, we 
113 
will outline the progress in hardware, software and 
114 
probe development as well as choices in biological 
115 
model and sample preparation that can help 
116 
improve live-cell SRM (Fig. 1) 
117 
Quantifying phototoxicity in microscopy 
118 
Measuring phototoxicity in microscopy is not a trivial 
119 
problem, as evidenced by the sparsity of the  
120 
 
121 
Fig. 2 Interactions of light with cellular components leading to 
122 
phototoxicity. a UV light can trigger apoptosis by inducing Fas 
123 
receptor-mediated signalling pathways. b UV light can directly 
124 
damage DNA by causing strand breakage (top) or thymidine 
125 
dimerisation (bottom), causing mutations and inducing DNA 
126 
damage responses. c UV and visible wavelengths can excite 
127 
photoactive molecules leading to chemical generation of ROS, 
128 
which can then damage cellular components. 
129 
available literature [12], [13]. This is not entirely 
130 
surprising, as phototoxicity is mediated by many 
131 
factors 
(Fig. 
1). 
These 
include 
illumination 
132 
wavelength, intensity and duration of illumination, 
133 
the illumination regime (e.g. LED illumination vs. 
134 
laser illumination, laser-scanning vs. light-sheet), 
135 
and the number of imaged 3D-planes [32]–[37]. 
136 
Additionally, 
illumination 
tolerance 
can 
vary 
137 
substantially between specimens (see Biological 
138 
Page 3 of 28
AUTHOR SUBMITTED MANUSCRIPT - JPhysD-122201.R1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Accepted Manuscript

----!@#$NewPage!@#$----
models and sample preparation section), and 
139 
experimental stress can influence a specimen’s 
140 
sensitivity to illumination [14]. For example, a 
141 
procedure as routine as transfection or the addition 
142 
of a drug has been shown to dramatically increase 
143 
cellular sensitivity to light [10], [38]. Therefore, steps 
144 
must be taken to reduce avoidable experimental 
145 
perturbations which can influence the well-being of 
146 
the sample in an illumination-independent manner, 
147 
e. g. 
suboptimal 
environmental 
conditions 
148 
(temperature, pH, etc.) [39] or complex sample 
149 
mounting. 
150 
How does one approach a problem as versatile as 
151 
measuring phototoxicity? An intuitive and common 
152 
way of assessing photodamage is by measuring 
153 
photobleaching [40]–[43]. However, phototoxicity 
154 
and photobleaching are two separate processes; 
155 
while 
toxic 
ROS 
are 
produced 
during 
156 
photobleaching, they can also be generated 
157 
independently of this process [15], [44]. Therefore, 
158 
phototoxicity can commence prior to a detectable 
159 
reduction in fluorescence, making photobleaching 
160 
an unreliable read-out for photodamage in the 
161 
context of live-cell imaging [12]. More importantly, 
162 
photobleaching rates give no information on the 
163 
health and viability of the specimen. Thus, a better 
164 
phototoxicity measure would have a read-out related 
165 
to the properties of the sample itself, rather than the 
166 
properties of the fluorescence [34]. 
167 
There are several in vitro assays for post-imaging 
168 
assessment of the health and viability of a specimen 
169 
that can be used to indicate whether phototoxicity 
170 
occurred (Fig. 3a). These include detection of toxic 
171 
ROS, fragmentation and oxidation of DNA strands, 
172 
reduced metabolic activity, loss of membrane 
173 
integrity and the expression of stress- and 
174 
apoptosis-related 
proteins 
[45]–[50]. 
The 
175 
advantages are that these assays provide an 
176 
inexpensive 
and 
simple 
specimen 
viability 
177 
evaluation. Thus, different illumination conditions 
178 
can be tested and viability can be assessed each 
179 
time. However, for such assays the measurement is 
180 
limited to a single timepoint and imaging cannot be 
181 
recommenced after performing the assay. 
182 
 
183 
Fig. 3 Methods for measuring phototoxicity. a `Destructive 
184 
read-outs' are techniques prohibiting further imaging of the 
185 
sample. These include blotting for phosphorylated forms of 
186 
proteins present in damage-activated pathways [51] and flow 
187 
cytometry for determining the population of cells expressing, 
188 
for example, apoptotic markers such as annexin V. b 
189 
`Fluorescent reporters' are additional indicators added to the 
190 
sample during imaging whose fluorescence signal changes in 
191 
response to e.g. intracellular Ca2+ concentration (top) or 
192 
mitochondrial membrane potential (bottom). `Label-free 
193 
methods' of quantifying phototoxicity involve: c short-term 
194 
observation of cell division and morphology and d proliferation 
195 
of cells in culture following imaging. 
196 
A more dynamic and practical approach entails 
197 
monitoring 
changes 
in 
relevant 
biological 
198 
parameters during imaging (Fig. 3b, c). Cellular 
199 
Page 4 of 28
AUTHOR SUBMITTED MANUSCRIPT - JPhysD-122201.R1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Accepted Manuscript

----!@#$NewPage!@#$----
processes which are particularly photosensitive (i.e. 
200 
rapidly perturbed by light) are excellent read-outs. 
201 
For example, a commonly employed method is 
202 
measuring 
changes 
in 
cytosolic 
calcium 
203 
concentration using calcium-sensitive fluorescent 
204 
probes [50], [52]–[54] (Fig. 3b, top). This strategy 
205 
was used to evaluate live-cell STED microscopy by 
206 
monitoring differences in intracellular calcium 
207 
concentration between control cells and STED-
208 
imaged cells. This method showed that while there 
209 
is little difference between calcium concentration in 
210 
control and STED-imaged cells when using 
211 
excitation and STED-lasers with wavelengths >600 
212 
nm, responses indicative of cell damage were 
213 
observed with shorter illumination wavelengths and 
214 
when longer STED-laser dwell times were used [29]. 
215 
Other processes exist that make suitable read-outs 
216 
for phototoxicity, including changes in mitochondrial 
217 
membrane potential [41], [51] (Fig. 3b, bottom), 
218 
reduction of chromosome movement [55] and 
219 
slowing of microtubule growth [10]. It is worth 
220 
highlighting that, regardless of the process chosen, 
221 
care must be taken when employing fluorescent 
222 
probes for visualising these read-outs [46], [56]. 
223 
There 
are 
image-based 
phototoxicity 
224 
measurements that can be performed without 
225 
fluorescent labels. These often rely on identifying 
226 
changes in cell morphology indicative of entry into 
227 
apoptosis, such as blebbing or cell rounding [10], 
228 
[14], [51], [57], for example by using transmitted light 
229 
imaging (Fig. 3c). This approach was recently used 
230 
to train a deep convolutional neural network, 
231 
referred to as ‘DeadNet’, with the objective to 
232 
automate phototoxicity detection and quantification 
233 
from transmitted light images [58]. However, despite 
234 
widespread use, relying on morphology as a read-
235 
out has two limitations: first, even experienced 
236 
researchers can struggle to identify subtle changes 
237 
in morphology, thus biasing the results (e.g. by 
238 
annotating ambiguous cases incorrectly [58]; 
239 
second, when changes become obvious, they 
240 
usually represent an extreme phenotype indicative 
241 
of irreversible damage. Thus, they cannot account 
242 
for early damage that may arise even as cells 
243 
display a healthy morphology [13], [39]. 
244 
In this context, a read-out that deserves special 
245 
mention is cell division (Fig. 3c, d): a well-
246 
characterised 
biological 
process 
with 
easily 
247 
identifiable phases. It is highly regulated and 
248 
sensitive 
to 
various 
perturbations, 
including 
249 
illumination and changes in ROS concentrations 
250 
[15], [31]. This makes cell cycle an excellent read-
251 
out for detection and quantification of phototoxicity 
252 
[39], with both continuous (Fig. 3c) and endpoint 
253 
(Fig. 3d) measurements possible. Delay in mitotic 
254 
progression has been used successfully to detect 
255 
perturbations in the health of both cultured cells and 
256 
developing 
embryos 
[32]–[35]. 
Additionally, 
257 
evaluating colony formation or number of cell 
258 
divisions after illumination (typically assessed after 
259 
a period of one or more cell cycles) can be indicative 
260 
of long-lasting damage [12], [29] (Fig. 3d). This 
261 
approach 
was 
used 
to 
perform 
extensive 
262 
characterisation of photodamage under illumination 
263 
conditions commonly used in single-molecule 
264 
localisation microscopy (SMLM) [10]. The viability of 
265 
several different cell lines was determined 20-24 h 
266 
post illumination, a strong correlation between 
267 
Page 5 of 28
AUTHOR SUBMITTED MANUSCRIPT - JPhysD-122201.R1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Accepted Manuscript

----!@#$NewPage!@#$----
shorter illumination wavelengths and increased cell 
268 
death was shown, particularly at high intensities. 
269 
However, results also suggested that long-term cell 
270 
viability 
is 
possible 
even 
with 
illumination 
271 
wavelengths as short as 405 nm, provided the 
272 
integrated light dose is small, preferably with 
273 
continuous 
rather 
than 
pulsed 
illumination. 
274 
Naturally, a limitation exists in employing these 
275 
methods to assess phototoxicity in post-mitotic 
276 
systems, e.g. primary neuron cultures. However, for 
277 
relevant models, choosing mitosis as a read-out has 
278 
the significant advantage of allowing phototoxicity 
279 
assessment based on label-free transmitted light 
280 
images [10], [29], [33], minimising the introducing 
281 
additional damage during evaluation. 
282 
From reports of phototoxicity in literature, several 
283 
conclusions can be drawn to guide live-cell friendly 
284 
SRM. Firstly, red-shifted wavelengths are preferable 
285 
to 
shorter 
wavelengths. 
In 
particular, 
UV 
286 
wavelengths should be avoided wherever possible 
287 
[10], [29], [33]. Furthermore, several studies 
288 
demonstrate that lower intensity illumination with 
289 
longer exposure is less damaging than short intense 
290 
bursts or pulses of illumination [10], [34], [40]. Most 
291 
importantly, a recurrent message throughout the 
292 
literature is that higher illumination intensities are 
293 
more 
damaging 
than 
corresponding 
imaging 
294 
conditions with lower illumination intensities. We 
295 
anticipate 
that 
real-time 
phototoxicity 
296 
measurements will become commonplace in both 
297 
diffraction-limited microscopy and SRM, and that 
298 
future SRM techniques will be accompanied by a 
299 
thorough description of how they impact living 
300 
Fig. 4 Low phototoxicity fluorescent probes and labelling for live-cell Super-Resolution Microscopy. Various recently-developed fluorescent protein- (a) and 
synthetic fluorophore- (b) based methods for labelling in live-cell super-resolution. All labels are shown attached to a microtubule as an example of an 
intracellular structure, with the exception of the Cer-HMSiR membrane dye in b. 
Page 6 of 28
AUTHOR SUBMITTED MANUSCRIPT - JPhysD-122201.R1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Accepted Manuscript

----!@#$NewPage!@#$----
samples. Concomitantly, for SRM users, awareness 
301 
of strategies for minimising phototoxicity is crucial. 
302 
Fluorescent probe development for live-cell 
303 
Super-Resolution Microscopy 
304 
SRM techniques have distinct requirements for 
305 
fluorescent probes. SIM quality relies on collecting 
306 
images of high Signal-to-Noise Ratio (SNR), 
307 
generally achieved by labelling with fluorophores of 
308 
high brightness and resistance to photobleaching. In 
309 
STED, fluorophores must not only be bright but also 
310 
possess a large Stokes-shift and stimulated 
311 
emission cross-section at the STED wavelength 
312 
[59]. SMLM techniques have the most demanding 
313 
labelling requirements - fluorophores must be 
314 
capable of cycling between ‘on’ and ‘off’ states with 
315 
appropriate kinetics, a high quantum yield in the on-
316 
state, and a very low quantum yield in the off-state. 
317 
Several fluorophores and probes have been 
318 
developed specifically for SRM [60], [61]. However, 
319 
while many specialised fluorophores exist for fixed 
320 
specimens [62], there are far fewer options available 
321 
for live-cell imaging. An inappropriate choice of 
322 
fluorophore for live-cell SRM will not only lead to low 
323 
quality images downstream [63], but also inevitably 
324 
impact acquisition settings and hence phototoxicity 
325 
[10], [64]. 
326 
As for most fluorescence microscopy techniques, 
327 
the two classes of fluorophores used in SRM are 
328 
fluorescent proteins (FPs) (Fig. 4a) and synthetic 
329 
fluorophores (SFs) (Fig. 4b). FPs are the usual 
330 
choice for live-cell imaging as they can be fused to 
331 
a target of interest via genetic encoding, but at the 
332 
cost of reduced brightness compared to SFs. The 
333 
recent development of bright and photobleaching-
334 
resistant FPs has expanded the options for SIM and 
335 
STED (Fig. 4a, left). Examples of these new FPs are 
336 
mNeonGreen 
(λex=506 nm) 
[65], 
mScarlet 
337 
(λex=569 nm) [66] and mGarnet (λex=598 nm) [67]. 
338 
SMLM 
techniques 
generally 
require 
339 
photoswitchable 
fluorophores 
(e.g. 
mEos3.2, 
340 
rsKame) [68], [69]. Despite the availability of several 
341 
photoswitchable FPs, their use in live-cell imaging 
342 
remains challenging [10], [64]. The chief reason is 
343 
that transitions between off- and on-states are 
344 
typically modulated by UV illumination. The 
345 
combination of this with high intensity excitation for 
346 
detection of molecular positions results in a short 
347 
window for live-cell SMLM studies. To reduce 
348 
phototoxicity in SMLM, FPs that do not require UV 
349 
pumping for photoswitching are being developed 
350 
(Fig. 4a, centre), with one such example being 
351 
SPOON [70]. Primed conversion is another 
352 
promising UV-independent approach to induce 
353 
photoswitching (Fig. 4a, right) [71]. Thereby a 
354 
combination of blue and near-infrared illumination 
355 
induces photoconversion in Dendra2 and the newly 
356 
developed primed-conversion protein pr-mEos2 
357 
[71], [72]. Recently, a general mechanism for primed 
358 
conversion was described, which is anticipated to 
359 
accelerate the development of more FPs that can be 
360 
photoconverted with this live-cell friendly approach 
361 
[73]. FPs for other specific SRM techniques have 
362 
also been developed (e.g. Skylan-NS for non-linear 
363 
SIM or GMars for REversible Saturable/switchable 
364 
OpticaL Fluorescence Transitions, RESOLFT) [74], 
365 
[75]. 
366 
Page 7 of 28
AUTHOR SUBMITTED MANUSCRIPT - JPhysD-122201.R1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Accepted Manuscript

----!@#$NewPage!@#$----
The second alternative, SFs (Fig. 4b), are small 
367 
chemically synthesised probes. These have higher 
368 
quantum yields and are more robust against 
369 
photobleaching than FPs [76]–[79].  While there are 
370 
some cell-permeable SFs that can be used to label 
371 
specific proteins (e.g. fluorogens such as SiR-
372 
tubulin and SiR-actin) (Fig. 4b, left) [80], [81] or cell 
373 
compartments directly (e.g. Membright, ER-Tracker 
374 
or MitoTracker) (Fig. 4b, centre) [77], [82]–[84], 
375 
additional `linker' molecules are normally required to 
376 
associate SFs with the structure of interest. These 
377 
linkers must bind the target structure with high 
378 
affinity 
and 
specificity 
(e.g. 
antibodies 
and 
379 
DNA/RNA scaffolds, usually using amine- or thiol-
380 
reactive derivatives of the SF) [85]. However, many 
381 
of these high-affinity linkers and SFs are not cell-
382 
permeable, which limits their use in live-cell SRM to 
383 
labelling of cell-surface molecules. If genetic 
384 
encoding is possible and preferable, cell-permeable 
385 
SFs can be combined with flexible self-labelling 
386 
systems, such as SNAP-tag, Halo-tag or FlAsH (Fig. 
387 
4b, right) [86]–[89]. An elegant example of such an 
388 
approach is the use of Cox8A-SNAP fusion labelled 
389 
with SNAP-Cell SiR for STED. This has enabled the 
390 
visualisation of the dynamics of mitochondrial 
391 
cristae with ~70 nm resolution [90]. 
392 
SFs have also been engineered for live-cell SRM. 
393 
Spontaneously blinking synthetic fluorophores (e.g. 
394 
HMSiR) have been recently developed (Fig. 4b, 
395 
center). They do not require UV irradiation or 
396 
cytotoxic additives (such as thiol) to induce 
397 
photoswitching [91], [92]. High photostability SFs 
398 
have also been developed, enabling live-cell STED 
399 
[79], [93]–[95]. 
400 
A final regime for live-cell SRM-compatible labelling 
401 
is based on site-specific conjugation of fluorophores 
402 
to a target of interest, through genetic code 
403 
modifications and click chemistry (Fig. 4b, right) 
404 
[96]–[98]. These approaches combine the benefits 
405 
of site-specific labelling (as is the case for FPs) with 
406 
no requirement for protein expression and bright 
407 
labels (as is the case for SFs). 
408 
 
409 
Biological models and sample preparation 
410 
Care should be taken when selecting a biological 
411 
model for SRM. Cellular sensitivity to light exposure 
412 
can vary based on cell type and species [10], [14], 
413 
[45], and in the case of whole organisms, 
414 
developmental stage [13], [34]. Phototoxicity has 
415 
been documented for different cell types, ranging 
416 
from primary cells [13], [45] to various immortalised 
417 
cell lines [10], [26], [38], [99]. One such study 
418 
focuses on immortalised cell lines, where it shows 
419 
that COS-7 and U2OS cells exhibit similar 
420 
photosensitivity, 
whereas 
HeLa 
cells 
are 
421 
substantially more robust, potentially making the 
422 
latter a more suitable system for live-cell SRM 
423 
studies [10]. Another study illustrated the effect of 
424 
photodamage on primary cells from rat central 
425 
nervous system [45]. Here, illumination with blue 
426 
light 
could 
induce 
morphological 
changes, 
427 
differentiation or cell death depending on the cell 
428 
type. 
429 
When 
imaging 
whole 
organisms, 
earlier 
430 
developmental stages from the same species tend 
431 
to be more photosensitive than later 
[12]. 
432 
Page 8 of 28
AUTHOR SUBMITTED MANUSCRIPT - JPhysD-122201.R1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Accepted Manuscript

----!@#$NewPage!@#$----
Furthermore, different model organisms display 
433 
variable photosensitivity. For example, fruit fly 
434 
embryos and nematode worms have higher 
435 
illumination tolerances than zebrafish embryos, 
436 
corals or cultured cells [13], [14]. Even within the 
437 
same cell, different intracellular structures exhibit 
438 
different responses to illumination [29], [100].  
439 
Photodamage can be mitigated through additional 
440 
sample preparation steps. Established strategies 
441 
centre on preventing photobleaching by modifying 
442 
the sample environment. As photobleaching can 
443 
contribute to phototoxicity via ROS production [44], 
444 
strategies to reduce photobleaching could also help 
445 
ameliorate phototoxicity [15], [29], [101]. One 
446 
strategy is to modify the environmental conditions 
447 
prior to or during imaging. A prime example is 
448 
removal 
of 
oxygen, 
the 
main 
effector 
of 
449 
photobleaching [102], from the culture medium. This 
450 
can be achieved by bubbling nitrogen through the 
451 
medium during imaging. This yields an increased 
452 
photostability [103], [104] and, since oxygen is 
453 
directly involved in the production of ROS, also 
454 
reduces light-dependent oxidative stress on the 
455 
sample. It has also been shown that growing cells in 
456 
a hypoxic environment (3% oxygen) yielded a 25% 
457 
increase in mitosis entry after blue light irradiation 
458 
[33]. Other approaches to reduce oxygen in the 
459 
medium involve the addition of commercially 
460 
available oxygen-scavengers such as the Oxyrase® 
461 
enzyme complex (developed by Oxyrase, Inc., 
462 
Mansfield, Ohio). In combination with suitable 
463 
substrates, such as D/L-lactate or D/L-succinate, 
464 
these 
enzymes 
catalytically 
reduce 
the 
465 
concentration of oxygen and free radicals present in 
466 
the medium, thus minimising photobleaching and 
467 
phototoxicity [105], [106]. While these approaches 
468 
could improve live-cell SRM, it should be noted that 
469 
they are only suitable for specimens which can 
470 
tolerate 
hypoxia 
or 
anoxia. 
Notably, 
some 
471 
fluorophores 
used 
in 
SRM 
require 
oxygen 
472 
scavenger systems to photoswitch, however, these 
473 
buffers typically use cytotoxic compounds such as 
474 
thiols, making them unsuitable for live-cell imaging. 
475 
A different strategy for reduction of ROS during 
476 
imaging involves supplementing the media with 
477 
antioxidants. Antioxidants are molecules that 
478 
prevent oxidation in a biological context [107]. 
479 
Among antioxidants, Trolox, the soluble form of 
480 
vitamin E, has been shown to have a protective 
481 
effect for a number of cell lines due to its ROS-
482 
neutralising properties [108]. The presence of the 
483 
antioxidant in the sample medium has been shown 
484 
to increase the number of post-illumination mitotic 
485 
cells by up to 38% compared to cells illuminated 
486 
without Trolox [33]. However, this molecule is not 
487 
suitable for SMLM, as it has been shown to inhibit 
488 
fluorophore blinking [109]. Another antioxidant used 
489 
in microscopy is rutin, a plant flavonoid shown to 
490 
reduce EGFP reddening [110], [111], although no 
491 
direct reduction of phototoxicity was demonstrated. 
492 
A notable example of a medium additive for live-cell 
493 
imaging is the vitamin- and antioxidant-rich 
494 
'Supplements for Optogenetic Survival' (SOS). SOS 
495 
has been shown to increase viability and reduce 
496 
photodamage in several cell types of the rat central 
497 
nervous system [45]. 
498 
There are chemicals used in mounting media, such 
499 
as various antioxidants, triplet-state quenchers and 
500 
Page 9 of 28
AUTHOR SUBMITTED MANUSCRIPT - JPhysD-122201.R1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Accepted Manuscript

----!@#$NewPage!@#$----
radical 
scavengers, 
that 
can 
be 
used 
for 
501 
photobleaching reduction and ROS neutralisation. 
502 
These include ascorbic acid [112], n-propyl gallate 
503 
[112]–[114], p-phenylenediamine [114]–[116], 1,4-
504 
diazobicyclo(2,2,2)-octane (DABCO) [114], [117], 
505 
mercaptoethylamine (MEA) and cyclooctatetraene 
506 
(COT) [112]. Their presence in mounting media for 
507 
reduction of photobleaching is well characterised 
508 
[112], 
[115], 
[118], 
however 
there 
is 
no 
509 
comprehensive study on the use of these chemicals 
510 
in live-cell imaging. As a result, there is little 
511 
information 
regarding 
biocompatible 
working 
512 
concentrations or biological side effects. Therefore, 
513 
while potentially useful, they require further 
514 
exploration prior to use in live-cell SRM.  
515 
Some substances commonly used as supplements 
516 
are known also to cause phototoxicity, such as 
517 
molecules with benzene rings which are intrinsically 
518 
fluorescent [111]. For example, common cell media 
519 
components, such as riboflavin and pyridoxal, can 
520 
enhance oxidative reddening of GFPs; this effect 
521 
accounts 
for 
a 
considerable 
part 
of 
GFP 
522 
photobleaching [119]. Depleting these substances 
523 
increases GFP photostability, indirectly reducing 
524 
photodamage [110]. Additionally, the combination of 
525 
riboflavin and tryptophan in media generates ROS 
526 
and induces cytotoxicity upon illumination, whereas 
527 
their removal alleviates this effect [120], [121]. 
528 
Finally, the study that established the SOS 
529 
supplement [45] used it in combination with the 
530 
photoinert media NEUMO and MEMO, which also 
531 
lack riboflavin. These media were specifically 
532 
developed to prevent phototoxicity of nervous 
533 
system cells. A confounding example is 4-(2-
534 
hydroxyethyl)-1-piperazineethanesulfonic 
acid 
535 
(HEPES), commonly used as a replacement for 
536 
carbon dioxide buffering during imaging [39]. 
537 
However, early reports demonstrated that HEPES-
538 
buffered media exposed to low-intensity white light 
539 
can 
generate 
toxic 
hydrogen 
peroxide 
with 
540 
detrimental effects on thymocyte or T-cell culture 
541 
[122], [123].  
542 
There is still a lack of information on SRM sample 
543 
preparation reducing phototoxicity. Many principles 
544 
can be transferred from conventional fluorescence 
545 
imaging. These include assessing photosensitivity 
546 
of the biological model, environmental conditions, 
547 
and attention to media composition. 
548 
Hardware developments for improved live-
549 
cell imaging 
550 
The microscope configuration has a substantial 
551 
impact on the amount of photodamage experienced 
552 
by a specimen. Fig. 5a shows the common 
553 
illumination regimes for conventional microscopy 
554 
and SRM (widefield for SIM and SMLM, confocal for 
555 
STED). Basic optimisations of the microscope body, 
556 
for example minimising photon loss in the detection 
557 
path by using high-quality filters and sensitive 
558 
detectors, will reduce the illumination burden to 
559 
achieve suitable SNR [101]. In SRM approaches, 
560 
microscopes are built with high-quality components, 
561 
often having bespoke solutions to maximize signal 
562 
detection [124], [125]. In addition, the ever-present 
563 
phototoxic high-intensity illumination requirements 
564 
of most SRM techniques can be further ameliorated 
565 
using dedicated hardware designs. Interestingly, a 
566 
recent study shown low-illumination live-cell SRM 
567 
Page 10 of 28
AUTHOR SUBMITTED MANUSCRIPT - JPhysD-122201.R1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Accepted Manuscript

----!@#$NewPage!@#$----
immediately followed by in situ fixation of the sample 
568 
and high-illumination SRM [126]. This approach 
569 
combines the collection of temporal information in 
570 
living-cells with a mild resolution increase, then 
571 
capture of higher resolution for a specific timepoint 
572 
upon fixation. 
573 
In the case of STED, the presence of a second high-
574 
intensity laser beam (depletion laser) in addition to a 
575 
confocal 
excitation 
beam 
confers 
the 
high 
576 
phototoxicity 
of 
this 
method. 
However, 
the 
577 
properties of both beams can have a substantial 
578 
impact on sample photodamage. It has been shown 
579 
in confocal microscopy that nanosecond pulsed, 
580 
rather than continuous, excitation can reduce 
581 
photobleaching, and that averaging multiple fast 
582 
scans is less phototoxic than acquiring a single slow 
583 
scan  (Fig. 5b, `Temporally adaptive illumination') 
584 
[41]. The properties of the excitation beam have also 
585 
been explored specifically in STED. For example, 
586 
reducing the pulsing rate of the excitation laser 
587 
allows time for long-lived triplet states to relax which 
588 
leads to decreased photobleaching [127]. Similarly 
589 
to confocal microscopy, scanning at a higher rate in 
590 
STED has been shown to reduce photobleaching 
591 
[42]; this is enabled by using fast resonant scanning 
592 
mirrors rather than slower galvanometer scanning 
593 
mirrors to scan the beam pair through the sample. 
594 
Another method described reducing phototoxicity in 
595 
STED is by using two-photon excitation (Fig. 5a 
596 
`Two-photon'). As two-photon excitation only excites 
597 
fluorophores within the focal volume of the beam 
598 
(rather than along the entire beam path, as is the 
599 
case in single-photon excitation), it is often 
600 
considered a more live-cell friendly imaging regime 
601 
[54], [128]. Indeed, live-cell STED has been 
602 
successfully 
demonstrated 
with 
two-photon 
603 
excitation [129], [130] although while the former 
604 
paper claims that there is no photodamage to the 
605 
sample, this is not quantified. It should be noted that 
606 
two-photon excitation does however increase local 
607 
heating, which can damage the sample in a non-
608 
fluorophore mediated manner [131].  
609 
Fig. 5 Hardware modalities for conventional and 
low-phototoxicity Super-Resolution Microscopy. a 
Microscopy illumination regimes for conventional 
fluorescence imaging. b Examples of regimes that 
reduce light dose to the sample by inhomogeneous 
illumination. c Examples of light-sheet microscopy 
geometries. 
Page 11 of 28
AUTHOR SUBMITTED MANUSCRIPT - JPhysD-122201.R1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Accepted Manuscript

----!@#$NewPage!@#$----
In STED microscopy with pulsed depletion lasers, 
610 
resolution scales non-linearly with beam intensity. 
611 
Thus, in order to obtain high resolution images, very 
612 
high (and phototoxic) depletion beam intensities are 
613 
required. A different approach to obtaining high 
614 
resolution STED images without this power 
615 
dependence is gSTED (gated-STED) [132]. gSTED 
616 
uses a continuous wave (CW) laser for the depletion 
617 
beam rather than a pulsed laser. When a CW 
618 
depletion beam is combined with a pulsed excitation 
619 
beam, spatial information about the underlying 
620 
fluorophore distribution becomes encoded in the 
621 
temporal information of emission on a nanosecond 
622 
timescale. By using time-gated detectors, photons 
623 
detected immediately after excitation can be 
624 
excluded from the final image, which improves 
625 
image resolution. By tuning the size of the time-gate, 
626 
gSTED can thus increase STED resolution 
627 
independent of increasing light dose to the sample 
628 
[133].  
629 
SIM is generally considered the least phototoxic 
630 
SRM technique [134]. However, it still requires the 
631 
acquisition of several frames (often ≥ 9) at high SNR 
632 
in order to generate the final reconstructed image. 
633 
Several approaches have been developed to reduce 
634 
the number of frames required for a SIM 
635 
reconstruction, including pixel reassignment and 
636 
image scanning microscopy (ISM) methods. One 
637 
example 
is 
multifocal 
structured 
illumination 
638 
microscopy 
(MSIM, 
[135]), 
which 
combines 
639 
principles from SIM and confocal microscopy to 
640 
scan an array of spots across the sample for fast 
641 
live-cell imaging with resolution doubling (Fig. 5b, 
642 
`Multi-focal illumination'). Another method, rapid 
643 
non-linear ISM [136], combines ISM with two-photon 
644 
excitation and second-harmonic generation for low 
645 
phototoxicity imaging. A wide range of such SIM-
646 
based techniques exist, and have been rigorously 
647 
compared elsewhere [134], [137]. It has been 
648 
demonstrated recently that using sub-millisecond 
649 
pulses as excitation in SIM (when combined with 
650 
novel analytics as described below) reduced 
651 
photobleaching and enables long-term live-cell 
652 
imaging [138]. 
653 
Techniques that restrict illumination to only the focal 
654 
plane of the sample are also preferable to those 
655 
which illuminate along the whole beam path. One 
656 
such example of this is TIRF (total internal reflection 
657 
fluorescence) microscopy, where only fluorophores 
658 
within a few hundred nanometers of the coverslip 
659 
are illuminated. While TIRF has been combined with 
660 
super-resolution modalities, such as SIM, and is 
661 
effective in reducing photodamage by axially 
662 
confining excitation [134], it is restrictive in that only 
663 
biological structures adjacent to the cell membrane 
664 
can be studied. 
665 
Light-sheet 
microscopy 
approaches 
similarly 
666 
confine illumination to a narrow band, but their 
667 
imaging geometries allow for investigation of 
668 
structures throughout the whole sample and not just 
669 
regions close to the coverslip. The majority of them 
670 
involve illuminating the sample with a thin sheet of 
671 
light 
and 
then 
detecting 
the 
fluorescence 
672 
perpendicular to the direction of sheet propagation 
673 
(Fig. 5c, `Gaussian light sheet') [139], [140]. This 
674 
confers low phototoxicity as only the part of the 
675 
sample being imaged is illuminated without the need 
676 
for non-linear optical processes (which is the case in 
677 
Page 12 of 28
AUTHOR SUBMITTED MANUSCRIPT - JPhysD-122201.R1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Accepted Manuscript

----!@#$NewPage!@#$----
two-photon 
microscopy). 
Indeed, 
light-sheet 
678 
microscopy was named the Nature Methods 
679 
technique of the year in 2014, in part due to its low 
680 
phototoxicity [141]. There are several ways in which 
681 
light-sheet microscopy schemes can yield super-
682 
resolution with reduced phototoxicity. Super-
683 
resolution in live samples has been demonstrated 
684 
using light-sheet microscopy by simply combining 
685 
this illumination geometry with SRM techniques 
686 
such as SMLM [142]–[144] and RESOLFT [145]. 
687 
However, the employed SRM methods still require 
688 
high-intensity illumination, and thus such composite 
689 
techniques do not exploit the inherent low 
690 
phototoxicity of light-sheet imaging. Therefore, a 
691 
more elegant approach involves illuminating the 
692 
sample with a light-sheet regime followed by the 
693 
application of SMLM analytics designed for ultra-
694 
high-density datasets, which allows for reduction of 
695 
the illumination power ([146] and Analytics section, 
696 
see below). The more widely-explored method for 
697 
combining SRM and light-sheet microscopy has 
698 
been the use of novel methods for generating and 
699 
shaping the light-sheet. Bessel beams have been 
700 
used to generate thinner light-sheets [147], and 
701 
these beams have also been extended to 
702 
incorporate SIM [148]. The latter strategy has also 
703 
been demonstrated on a system with two 
704 
counterpropagating 
light-sheets 
formed 
using 
705 
standard Gaussian beams [149]. The most radical 
706 
and live-imaging-friendly light-sheet SRM technique 
707 
developed to date is lattice light-sheet microscopy 
708 
[150] (Fig. 5c, `Lattice light sheet'). This has 
709 
demonstrated 
3D 
time-lapse 
super-resolution 
710 
imaging in both cultured cells and intact model 
711 
organisms with minimal phototoxicity. 
712 
An interesting approach to reducing the illumination 
713 
dose in SRM is using spatially varying illumination 
714 
depending on the structural content of the imaging 
715 
region (Fig. 5b, `Spatially adaptive illumination'). 
716 
This approach was originally demonstrated for 
717 
Fig. 6 Analytics to complement low-phototoxicity 
imaging regimes. a Top: typical SMLM images are 
successfully reconstructed from sparse blinking raw 
data acquired under high phototoxic illumination. 
Bottom: reducing phototoxic illumination leads to more 
emitting fluorophores per raw data frame. When 
reconstructed using conventional SMLM algorithms, 
these produce low-quality images containing artefacts. 
High density SMLM algorithms can produce better 
quality images from such datasets. b Top: typical SIM 
imaging involves acquiring 9-25 raw images (depending 
on the number of grating rotations and phases) at high 
SNR, which can be successfully reconstructed using 
conventional SIM algorithms. Bottom: decreasing the 
illumination intensity, and thus SNR of the raw images, 
leads to artefacts in images reconstructed using 
conventional methods. The Hessian SIM deconvolution 
algorithm can bypass this limitation [138]. c Deep neural 
networks can be trained to infer super-resolution 
information from e.g. low-resolution diffraction-limited 
or low-quality super-resolution images. In this example, 
a neural network can be trained on pairs of low 
resolution/super-resolution images of the trained 
structure (`Network training'). The trained network can 
then be applied to unseen low resolution images to infer 
the super-resolution equivalents (`Network inference'). 
Page 13 of 28
AUTHOR SUBMITTED MANUSCRIPT - JPhysD-122201.R1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Accepted Manuscript

----!@#$NewPage!@#$----
confocal imaging [48] and has since been extended 
718 
to SIM [151], RESOLFT [152] and indeed light-sheet 
719 
microscopy [141]. There is also a range of adaptive 
720 
illumination STED techniques that have been 
721 
developed 
[153]–[155], 
and 
while 
these 
722 
predominantly focus on reducing light dose in the 
723 
context of photobleaching, this will concomitantly 
724 
also impact the live-cell compatibility of these 
725 
techniques. 
726 
Analytical approaches to live-cell Super-
727 
Resolution Microscopy 
728 
Analytics can be used to extract super-resolution 
729 
information 
from 
images 
acquired 
at 
low 
730 
illumination, and thus low phototoxicity (Fig. 6). Such 
731 
techniques are generally based on SMLM principles 
732 
but improve its live-cell compatibility (Fig. 6a). In 
733 
SMLM, when high intensity illumination is used, 
734 
fluorophore blinking is sparse and thus the well-
735 
separated single molecules are straightforward to 
736 
detect and localise with high accuracy and precision 
737 
[156], [157]. However, as intensity is decreased 
738 
towards a lower phototoxicity regime, blinking 
739 
becomes more dense and molecules become 
740 
increasingly overlapped. Such datasets require 
741 
specialised algorithms to extract molecule locations. 
742 
The first example of such an algorithm was Super-
743 
Resolution Optical Fluctuation imaging (SOFI), 
744 
where the temporal statistics of fluorophore intensity 
745 
oscillations are used to generate images with sub-
746 
diffraction resolution [158]. Indeed, SOFI has been 
747 
used to image live cells [159] although only for short 
748 
periods of time due to the requirement for UV 
749 
illumination to induce photoswitching. Another 
750 
algorithm developed for analysing datasets with 
751 
dense blinking is 3B [160], where super-resolution 
752 
images can be obtained from datasets imaged with 
753 
a xenon arc lamp rather than lasers. However, both 
754 
SOFI 
and 
3B 
techniques 
still 
rely 
on 
755 
photoswitchable 
fluorophores, 
which 
have 
756 
drawbacks discussed above. The Super-Resolution 
757 
Radial Fluctuations (SRRF) algorithm allows for the 
758 
reconstruction of super-resolution images from 
759 
datasets 
containing 
non-photoswitchable 
760 
fluorophores such as GFP [161], [162]. SRRF has 
761 
been shown to work on datasets obtained with 
762 
confocal and LED-illuminated microscopes, with the 
763 
latter enabling continuous live-cell imaging for >30 
764 
minutes [163]. However, SRRF cannot retrieve 
765 
resolutions in these regimes as high as those 
766 
achievable with photoswitchable fluorophores. A 
767 
promising new development for analysing high-
768 
density datasets is Haar wavelet kernel (HAWK) 
769 
[164]. HAWK is a pre-processing algorithm that 
770 
separates fluorophores in time; this creates an 
771 
artificial lower-density dataset, which can then be 
772 
analysed using any SMLM algorithm. 
773 
While most analytical developments for live-cell 
774 
SRM centre on SMLM-based techniques, there are 
775 
also analytics for enabling lower phototoxicity 
776 
imaging in SIM and STED. Hessian-SIM is a 
777 
deconvolution algorithm that can obtain high-quality 
778 
SIM images from raw data acquired at low signal-to-
779 
noise ratio  (Fig. 6b) [138]. This overcomes a 
780 
substantial barrier in SIM, in that conventional SIM 
781 
reconstruction algorithms perform poorly on low-
782 
illumination datasets, leading to artefacts within the 
783 
resulting images. Approaches have also been 
784 
Page 14 of 28
AUTHOR SUBMITTED MANUSCRIPT - JPhysD-122201.R1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Accepted Manuscript

----!@#$NewPage!@#$----
proposed for low-power STED microscopy based on 
785 
reconstructing 
images 
with 
knowledge 
of 
786 
fluorescence lifetime changes induced by the STED 
787 
beam [75], [165].  
788 
A rapidly evolving field in microscopy image analysis 
789 
is the use of machine learning (ML)-based 
790 
techniques [166], [167]. Such techniques are used 
791 
for 
diverse 
applications 
including 
object 
792 
segmentation, denoising, and structure prediction, 
793 
and these can also be extended to SRM (Fig. 6c). 
794 
One example is Content Aware Image Restoration 
795 
(CARE), where a neural network is trained on high 
796 
illumination 
intensity 
datasets 
(i.e. 
high 
797 
phototoxicity) and used to denoise corresponding 
798 
datasets acquired at much lower illumination 
799 
intensities [168]. CARE was demonstrated to 
800 
enhance resolution of GFP-tagged microtubules to 
801 
a similar extent to SRRF analysis of the same data, 
802 
but with higher quality and higher temporal 
803 
resolution. There are also specialised ML algorithms 
804 
for super-resolution applications. ANNA-PALM is a 
805 
method that, after training a neural network with 
806 
sparse SMLM data, can 
reconstruct super-
807 
resolution images from dense data and a 
808 
correspondingly lower number of frames [169]. 
809 
While not demonstrated in live-cell data, this 
810 
technique could in theory alleviate phototoxicity with 
811 
minimal sacrifice to spatial resolution by imaging 
812 
photoswitchable 
FPs 
with 
lower 
illumination 
813 
intensity. Other ML-based techniques have also 
814 
allowed for prediction of enhanced resolution 
815 
images from low illumination diffraction-limited 
816 
images (Fig. 6c), for example: converting confocal 
817 
to Airyscan-type or STED-type images [75], [170]; or 
818 
widefield to SIM-type images [75]. 
819 
Discussion and outlook 
820 
High quality live-cell fluorescence microscopy 
821 
involves compromising between four key properties: 
822 
SNR, imaging speed, spatial resolution, and sample 
823 
health [12]. We present an overview of the 
824 
challenges faced on how to balance the latter two 
825 
properties in live-cell SRM, highlighting potential 
826 
strategies to maximise resolution while minimising 
827 
phototoxicity. 
828 
As commercial super-resolution systems become 
829 
commonplace in biological labs and open-source 
830 
microscope hardware becomes more widespread, 
831 
there is a growing desire to translate cell biology 
832 
experiments from conventional diffraction-limited 
833 
microscopes to higher resolution alternatives. 
834 
However, the cost of this increased resolution is 
835 
often the sample health. Users must be aware of 
836 
what phototoxicity is, how to detect it, and methods 
837 
that can be used to ameliorate it. Unfortunately, 
838 
there are very few dedicated studies discussing 
839 
phototoxicity specifically in SRM [10], [29]. 
840 
It is clear that there are several frontiers for 
841 
optimising 
SRM 
protocols 
for 
minimising 
842 
phototoxicity, and a much-needed development in 
843 
the field is a non-perturbing robust indicator of 
844 
sample health during imaging. Caution must be 
845 
taken when reporting and evaluating phototoxicity 
846 
as it would also require using uniform metrics for 
847 
data quality. There is already software available for 
848 
assessing the quality and resolution of SRM images 
849 
[171], [172] Comparative analytics for phototoxicity 
850 
Page 15 of 28
AUTHOR SUBMITTED MANUSCRIPT - JPhysD-122201.R1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Accepted Manuscript

----!@#$NewPage!@#$----
would thus provide a complete numerical framework 
851 
for experiment optimisation. 
852 
As 
super-resolution 
microscopes 
become 
853 
increasingly standard equipment in biological 
854 
research, users must be aware of their limitations in 
855 
live-cell imaging. Many of the suggestions offered in 
856 
this review for reducing phototoxicity remain under 
857 
active development, and it is imperative for users to 
858 
follow 
progress 
in 
hardware, 
analytics 
and 
859 
fluorophores to ensure that they are minimising 
860 
photodamage to samples. 
861 
Acknowledgements 
862 
This work was funded by grants from the UK 
863 
Biotechnology and Biological Sciences Research 
864 
Council 
(BB/R000697/1; 
BB/R021805/1; 
865 
BB/S507532/1) (R.H.), the UK Medical Research 
866 
Council (MR/K015826/1) (R.H.), the Wellcome Trust 
867 
(203276/Z/16/Z) (S.C., R.H). K.L.T. supported by a 
868 
4-year MRC Research Studentship and Y.Y. by a 4-
869 
year BBSRC Research Studentship. We would like 
870 
to thank Dr. David Albrecht (Max Planck Institute for 
871 
the Science of Light, Erlangen, Germany), Dr. 
872 
Francesca Bottanelli (Freie Universität Berlin, Berlin, 
873 
Germnay), Dr. Agathe Chaigne (UCL, London, UK), 
874 
Dr. Gautam Dey (UCL, London, UK), Dr. Caron 
875 
Jacobs (University of Cape Town, Cape Town, 
876 
South Africa), Ms. Megan Jones (UCL, London, UK), 
877 
Dr. Christophe Leterrier (Aix Marseille Université, 
878 
Marseille, France), Dr. Apostolos Papandreou (UCL, 
879 
London, UK) and Dr. Uwe Schmidt (Max Planck 
880 
Institute of Molecular Cell Biology and Genetics, 
881 
Dresden, Germany) for valuable advice and 
882 
feedback. 
883 
 
884 
ORCID IDs 
885 
Kalina L. Tosheva: 
886 
https://orcid.org/0000-0003-0999-5182 
887 
Yue Yuan: 
888 
https://orcid.org/0000-0002-6698-3009 
889 
Pedro Matos Pereira:  
890 
https://orcid.org/0000-0002-1426-9540 
891 
Siân Culley: 
892 
https://orcid.org/0000-0003-2112-0143 
893 
Ricardo Henriques:  
894 
https://orcid.org/0000-0002-2043-5234 
895 
Page 16 of 28
AUTHOR SUBMITTED MANUSCRIPT - JPhysD-122201.R1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Accepted Manuscript

----!@#$NewPage!@#$----
Bibliography 
896 
[1] 
L. Schermelleh et al., “Super-resolution microscopy demystified,” Nat. Cell Biol., vol. 21, no. 1, pp. 72–84, 
897 
2019. 
898 
[2] 
E. Betzig et al., “Imaging Intracellular Fluorescent Proteins at Nanometer Resolution,” Science (80-. )., 
899 
vol. 313, no. 5793, pp. 1642–1645, 2006. 
900 
[3] 
M. J. Rust, M. Bates, and X. Zhuang, “Sub-diffraction-limit imaging by stochastic optical reconstruction 
901 
microscopy (STORM),” Nat. Methods, vol. 3, no. 10, pp. 793–796, 2006. 
902 
[4] 
M. G. L. Gustafsson, “Surpassing the lateral resolution limit by a factor of two using structured 
903 
illumination microscopy,” J. Microsc., vol. 198, no. 2, pp. 82–87, May 2000. 
904 
[5] 
S. W. Hell and J. Wichmann, “Breaking the diffraction resolution limit by stimulated emission: stimulated-
905 
emission-depletion fluorescence microscopy,” Opt. Lett., vol. 19, no. 11, pp. 780–782, Jun. 1994. 
906 
[6] 
K. Xu, G. Zhong, and X. Zhuang, “Actin, Spectrin, and Associated Proteins Form a Periodic Cytoskeletal 
907 
Structure in Axons,” Science (80-. )., vol. 339, no. 6118, pp. 452 LP – 456, Jan. 2013. 
908 
[7] 
J. Nixon-Abell et al., “Increased spatiotemporal resolution reveals highly dynamic dense tubular matrices 
909 
in the peripheral ER,” Science (80-. )., vol. 354, no. 6311, p. aaf3928, Oct. 2016. 
910 
[8] 
J. Chojnacki et al., “Envelope glycoprotein mobility on HIV-1 particles depends on the virus maturation 
911 
state,” Nat. Commun., vol. 8, no. 1, p. 545, 2017. 
912 
[9] 
T. Nerreter et al., “Super-resolution microscopy reveals ultra-low CD19 expression on myeloma cells that 
913 
triggers elimination by CD19 CAR-T,” Nat. Commun., vol. 10, no. 1, p. 3137, 2019. 
914 
[10] 
S. Wäldchen, J. Lehmann, T. Klein, S. van de Linde, and M. Sauer, “Light-induced cell damage in live-
915 
cell super-resolution microscopy,” Sci. Rep., vol. 5, p. 15348, Oct. 2015. 
916 
[11] 
R. Henriques, C. Griffiths, E. Hesper Rego, and M. M. Mhlanga, “PALM and STORM: Unlocking live-cell 
917 
super-resolution,” Biopolymers, vol. 95, no. 5, pp. 322–331, May 2011. 
918 
[12] 
P. P. Laissue, R. A. Alghamdi, P. Tomancak, E. G. Reynaud, and H. Shroff, “Assessing phototoxicity in 
919 
live fluorescence imaging,” Nat. Methods, vol. 14, p. 657, Jun. 2017. 
920 
[13] 
J. Icha, M. Weber, J. C. Waters, and C. Norden, “Phototoxicity in live fluorescence microscopy, and how 
921 
to avoid it,” BioEssays, vol. 39, no. 8, p. 1700003, Aug. 2017. 
922 
[14] 
V. Magidson and A. Khodjakov, Circumventing photodamage in live-cell microscopy, 4th ed., vol. 114. 
923 
Elsevier Inc., 2013. 
924 
[15] 
R. Dixit and R. Cyr, “Cell damage and reactive oxygen species production induced by fluorescence 
925 
microscopy: effect on mitosis and guidelines for non-invasive fluorescence microscopy,” Plant J., vol. 36, 
926 
no. 2, pp. 280–290, 2003. 
927 
[16] 
K. Logg, K. Bodvard, A. Blomberg, and M. Käll, “Investigations on light-induced stress in fluorescence 
928 
microscopy using nuclear localization of the transcription factor Msn2p as a reporter,” FEMS Yeast Res., 
929 
Page 17 of 28
AUTHOR SUBMITTED MANUSCRIPT - JPhysD-122201.R1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Accepted Manuscript

----!@#$NewPage!@#$----
vol. 9, no. 6, pp. 875–884, 2009. 
930 
[17] 
Y. Devary, C. Rosette, J. A. DiDonato, and M. Karin, “NF-kappa B activation by ultraviolet light not 
931 
dependent on a nuclear signal,” Science (80-. )., vol. 261, no. 5127, pp. 1442–1445, 1993. 
932 
[18] 
A. Rehemtulla, C. A. Hamilton, A. M. Chinnaiyan, and V. M. Dixit, “Ultraviolet Radiation-induced 
933 
Apoptosis Is Mediated by Activation of CD-95 (Fas/APO-1),” J. Biol. Chem. , vol. 272, no. 41, pp. 25783–
934 
25786, Oct. 1997. 
935 
[19] 
M. J. Peak, J. G. Peak, and B. A. Carnes, “Induction of Direct and Indirect Single‐Strand Breaks in 
936 
Human Cell Dna By Far‐ and Near‐Ultraviolet Radiations: Action Spectrum and Mechanisms,” 
937 
Photochem. Photobiol., vol. 45, no. 3, pp. 381–387, 1987. 
938 
[20] 
M. J. Peak and J. G. Peak, “SINGLE-STRAND BREAKS INDUCED IN BACILLUS SUBTILIS DNA BY 
939 
ULTRAVIOLET LIGHT: ACTION SPECTRUM and PROPERTIES,” Photochem. Photobiol., vol. 35, no. 5, 
940 
pp. 675–680, May 1982. 
941 
[21] 
B. Durbeej and L. A. Eriksson, “Reaction mechanism of thymine dimer formation in DNA induced by UV 
942 
light,” J. Photochem. Photobiol. A Chem., vol. 152, no. 1, pp. 95–101, 2002. 
943 
[22] 
D. E. Brash et al., “A role for sunlight in skin cancer: UV-induced p53 mutations in squamous cell 
944 
carcinoma.,” Proc. Natl. Acad. Sci., vol. 88, no. 22, pp. 10124–10128, 1991. 
945 
[23] 
H. Vrieling et al., “DNA strand specificity for UV-induced mutations in mammalian cells.,” Mol. Cell. Biol., 
946 
vol. 9, no. 3, pp. 1277–1283, 1989. 
947 
[24] 
M. L. Cunningham, J. S. Johnson, S. M. Giovanazzi, and M. J. Peak, “Photosensitized Production of 
948 
Superoxide Anion by Monochromatic (290–405 nm) Ultraviolet Irradiation of NADH and NADPH 
949 
Coenzymes,” Photochem. Photobiol., vol. 42, no. 2, pp. 125–128, 1985. 
950 
[25] 
M. Eichler, R. Lavi, A. Shainberg, and R. Lubart, “Flavins are source of visible-light-induced free radical 
951 
formation in cells,” Lasers Surg. Med., vol. 37, no. 4, pp. 314–319, Oct. 2005. 
952 
[26] 
P. E. Hockberger et al., “Activation of flavin-containing oxidases underlies light-induced production of 
953 
H2O2 in mammalian cells,” Proc. Natl. Acad. Sci., vol. 96, no. 11, pp. 6255–6260, 1999. 
954 
[27] 
G. Y. Fraikin, M. G. Strakhovskaya, and A. B. Rubin, “The role of membrane-bound porphyrin-type 
955 
compound as endogenous sensitizer in photodynamic damage to yeast plasma membranes,” J. 
956 
Photochem. Photobiol. B Biol., vol. 34, no. 2–3, pp. 129–135, 1996. 
957 
[28] 
F. Ricchelli, “Photophysical properties of porphyrins in biological membranes,” J. Photochem. Photobiol. 
958 
B Biol., vol. 29, no. 2, pp. 109–118, 1995. 
959 
[29] 
N. Kilian et al., “Assessing photodamage in live-cell STED microscopy,” Nat. Methods, vol. 15, no. 10, pp. 
960 
755–756, 2018. 
961 
[30] 
C. S. Foote, “DEFINITION OF TYPE I and TYPE II PHOTOSENSITIZED OXIDATION,” Photochem. 
962 
Photobiol., vol. 54, no. 5, p. 659, Nov. 1991. 
963 
Page 18 of 28
AUTHOR SUBMITTED MANUSCRIPT - JPhysD-122201.R1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Accepted Manuscript

----!@#$NewPage!@#$----
[31] 
W. C. Burhans and N. H. Heintz, “The cell cycle is a redox cycle: Linking phase-specific targets to cell 
964 
fate,” Free Radic. Biol. Med., vol. 47, no. 9, pp. 1282–1293, 2009. 
965 
[32] 
Z. Schilling et al., “Predictive-focus illumination for reducing photodamage in live-cell microscopy,” J. 
966 
Microsc., vol. 246, no. 2, pp. 160–167, May 2012. 
967 
[33] 
S. Douthwright and G. Sluder, “Live Cell Imaging: Assessing the Phototoxicity of 488 and 546 nm Light 
968 
and Methods to Alleviate it,” J. Cell. Physiol., vol. 232, no. 9, pp. 2461–2468, 2017. 
969 
[34] 
J.-Y. Tinevez et al., “Chapter fifteen - A Quantitative Method for Measuring Phototoxicity of a Live Cell 
970 
Imaging Microscope,” in Imaging and Spectroscopic Analysis of Living Cells, vol. 506, P. M. B. T.-M. in E. 
971 
Conn, Ed. Academic Press, 2012, pp. 291–309. 
972 
[35] 
P. M. Carlton et al., “Fast live simultaneous multiwavelength four-dimensional optical microscopy,” Proc. 
973 
Natl. Acad. Sci., vol. 107, no. 37, pp. 16016–16022, 2010. 
974 
[36] 
M. M. Knight, S. R. Roberts, D. A. Lee, and D. L. Bader, “Live cell imaging using confocal microscopy 
975 
induces  intracellular calcium transients and cell death,” Am. J. Physiol. Physiol., vol. 284, no. 4, pp. 
976 
C1083–C1089, Apr. 2003. 
977 
[37] 
Y. Wu et al., “Inverted selective plane illumination microscopy (iSPIM) enables coupled cell identity 
978 
lineaging and neurodevelopmental imaging in Caenorhabditis elegans,” Proc. Natl. Acad. Sci., vol. 108, 
979 
no. 43, pp. 17708–17713, 2011. 
980 
[38] 
H. Schneckenburger et al., “Light exposure and cell viability in fluorescence microscopy,” J. Microsc., vol. 
981 
245, no. 3, pp. 311–318, 2012. 
982 
[39] 
R. Cole, “Live-cell imaging,” Cell Adh. Migr., vol. 8, no. 5, pp. 452–459, Sep. 2014. 
983 
[40] 
F. Mubaid and C. M. Brown, “Less is More: Longer Exposure Times with Low Light Intensity is Less 
984 
Photo-Toxic,” Micros. Today, vol. 25, no. 6, pp. 26–35, 2017. 
985 
[41] 
C. Boudreau, T. L. Wee, Y. R. Duh, M. P. Couto, K. H. Ardakani, and C. M. Brown, “Excitation light dose 
986 
engineering to reduce photo-bleaching and photo-toxicity,” Sci. Rep., vol. 6, pp. 1–12, 2016. 
987 
[42] 
Y. Wu, X. Wu, R. Lu, J. Zhang, L. Toro, and E. Stefani, “Resonant Scanning with Large Field of View 
988 
Reduces Photobleaching and Enhances Fluorescence Yield in STED Microscopy,” Sci. Rep., vol. 5, p. 
989 
14766, Oct. 2015. 
990 
[43] 
E. A. Halabi, D. Pinotsi, and P. Rivera-Fuentes, “Photoregulated fluxional fluorophores for live-cell super-
991 
resolution microscopy with no apparent photobleaching,” Nat. Commun., vol. 10, no. 1, p. 1232, 2019. 
992 
[44] 
K. Jacobson, Z. Rajfur, E. Vitriol, and K. Hahn, “Chromophore-assisted laser inactivation in cell biology,” 
993 
Trends Cell Biol., vol. 18, no. 9, pp. 443–450, 2008. 
994 
[45] 
J. H. Stockley et al., “Surpassing light-induced cell damage in vitro with novel cell culture media,” Sci. 
995 
Rep., vol. 7, no. 1, p. 849, 2017. 
996 
[46] 
J. W. Dobrucki, D. Feret, and A. Noatynska, “Scattering of exciting light by live cells in fluorescence 
997 
Page 19 of 28
AUTHOR SUBMITTED MANUSCRIPT - JPhysD-122201.R1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Accepted Manuscript

----!@#$NewPage!@#$----
confocal imaging: Phototoxic effects and relevance for FRAP studies,” Biophys. J., vol. 93, no. 5, pp. 
998 
1778–1786, 2007. 
999 
[47] 
C. Roechlecke, A. Schaller, L. Knels, and R. H. W. Funk, “The Influence of Sublethal Blue Light Exposure 
1000 
on Human {{RPE}} Cells,” Mol. Vis., vol. 15, pp. 1929–1938, 2009. 
1001 
[48] 
R. A. Hoebe, C. H. Van Oven, T. W. J. Gadella, P. B. Dhonukshe, C. J. F. Van Noorden, and E. M. M. 
1002 
Manders, “Controlled light-exposure microscopy reduces photobleaching and phototoxicity in 
1003 
fluorescence live-cell imaging,” Nat. Biotechnol., vol. 25, no. 2, pp. 249–253, 2007. 
1004 
[49] 
J. Ge et al., “Standard fluorescent imaging of live cells is highly genotoxic,” Cytom. Part A, vol. 83A, no. 
1005 
6, pp. 552–560, Jun. 2013. 
1006 
[50] 
D. H. Hawkins and H. Abrahamse, “The role of laser fluence in cell viability, proliferation, and membrane 
1007 
integrity of wounded human skin fibroblasts following helium-neon laser irradiation,” Lasers Surg. Med., 
1008 
vol. 38, no. 1, pp. 74–83, Jan. 2006. 
1009 
[51] 
Y. Kuse, K. Ogawa, K. Tsuruma, M. Shimazawa, and H. Hara, “Damage of photoreceptor-derived cells in 
1010 
culture induced by light emitting diode-derived blue light,” Sci. Rep., vol. 4, pp. 1–12, 2014. 
1011 
[52] 
A. McDonald, J. Harris, D. MacMillan, J. Dempster, and G. McConnell, “Light-induced Ca^2+ transients 
1012 
observed in widefield epi-fluorescence microscopy of excitable cells,” Biomed. Opt. Express, vol. 3, no. 6, 
1013 
p. 1266, 2012. 
1014 
[53] 
H. J. Koester, D. Baur, R. Uhl, and S. W. Hell, “Ca2+ fluorescence imaging with pico- and femtosecond 
1015 
two-photon excitation: Signal and photodamage,” Biophys. J., vol. 77, no. 4, pp. 2226–2236, 1999. 
1016 
[54] 
A. Hopt and E. Neher, “Highly Nonlinear Photodamage in Two-Photon Fluorescence Microscopy,” 
1017 
Biophys. J., vol. 80, no. 4, pp. 2029–2036, 2001. 
1018 
[55] 
C.-H. Chuang, A. E. Carpenter, B. Fuchsova, T. Johnson, P. de Lanerolle, and A. S. Belmont, “Long-
1019 
Range Directional Movement of an Interphase Chromosome Site,” Curr. Biol., vol. 16, no. 8, pp. 825–
1020 
831, 2006. 
1021 
[56] 
T. Minamikawa, D. A. Williams, D. N. Bowser, and P. Nagley, “Mitochondrial Permeability Transition and 
1022 
Swelling Can Occur Reversibly without Inducing Cell Death in Intact Human Cells,” Exp. Cell Res., vol. 
1023 
246, no. 1, pp. 26–37, 1999. 
1024 
[57] 
R. A. Hoebe, H. T. M. van der Voort, J. Stap, C. J. F. van Noorden, and E. M. M. Mamders, “Quantitative 
1025 
determination of the reduction of phototoxicity and photobleaching by controlled light exposure 
1026 
microscopy,” J. Microsc., vol. 231, no. 1, pp. 9–20, Jul. 2008. 
1027 
[58] 
D. Richmond, A. P.-T. Jost, T. Lambert, J. Waters, and H. Elliott, “DeadNet: Identifying Phototoxicity from 
1028 
Label-free Microscopy Images of Cells using Deep ConvNets,” CoRR, vol. abs/1701.0, 2017. 
1029 
[59] 
G. Vicidomini, P. Bianchini, and A. Diaspro, “STED super-resolved microscopy,” Nat. Methods, vol. 15, 
1030 
no. 3, pp. 173–182, 2018. 
1031 
Page 20 of 28
AUTHOR SUBMITTED MANUSCRIPT - JPhysD-122201.R1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Accepted Manuscript

----!@#$NewPage!@#$----
[60] 
R. Henriques and M. M. Mhlanga, “PALM and STORM: What hides beyond the Rayleigh limit?,” 
1032 
Biotechnol. J., vol. 4, no. 6, pp. 846–857, Jun. 2009. 
1033 
[61] 
J. C. Chozinski, Tyler J., Gagnon, Lauren A., Vaughan, “Twinkle, twinkle little star: photoswitchable 
1034 
fluorophores for super-resolution imaging,” FEBS Lett., vol. 588, p. 3603, 2014. 
1035 
[62] 
G. T. Dempsey, J. C. Vaughan, K. H. Chen, M. Bates, and X. Zhuang, “Evaluation of fluorophores for 
1036 
optimal performance in localization-based super-resolution imaging,” Nat. Methods, vol. 8, no. 12, pp. 
1037 
1027–1040, 2011. 
1038 
[63] 
J. V. Thevathasan et al., “Nuclear pores as versatile reference standards for quantitative superresolution 
1039 
microscopy,” bioRxiv, 2019. 
1040 
[64] 
S. van de Linde, I. Krstić, T. Prisner, S. Doose, M. Heilemann, and M. Sauer, “Photoinduced formation of 
1041 
reversible dye radicals and their impact on super-resolution imaging.,” Photochem. Photobiol. Sci., vol. 
1042 
10, no. 4, pp. 499–506, 2011. 
1043 
[65] 
N. C. Shaner et al., “A bright monomeric green fluorescent protein derived from Branchiostoma 
1044 
lanceolatum,” Nat. Methods, vol. 10, no. 5, pp. 407–409, 2013. 
1045 
[66] 
D. S. Bindels et al., “mScarlet: a bright monomeric red fluorescent protein for cellular imaging,” Nat. 
1046 
Methods, vol. 14, no. 1, pp. 53–56, 2017. 
1047 
[67] 
A. Hense, B. Prunsche, P. Gao, Y. Ishitsuka, K. Nienhaus, and G. U. Nienhaus, “Monomeric Garnet, a 
1048 
far-red fluorescent protein for live-cell STED imaging,” Sci. Rep., vol. 5, pp. 1–10, 2015. 
1049 
[68] 
M. Zhang et al., “Rational design of true monomeric and bright photoactivatable fluorescent proteins,” 
1050 
Nat. Methods, vol. 9, no. 7, pp. 727–729, 2012. 
1051 
[69] 
A. B. Rosenbloom, S.-H. S.-H. Lee, M. To, A. Lee, J. Y. Shin, and C. Bustamante, “Optimized two-color 
1052 
super resolution imaging of Drp1 during mitochondrial fission with a slow-switching Dronpa variant,” Proc. 
1053 
Natl. Acad. Sci., vol. 111, no. 36, pp. 13093–13098, 2014. 
1054 
[70] 
Y. Arai et al., “Spontaneously Blinking Fluorescent Protein for Simple Single Laser Super-Resolution Live 
1055 
Cell Imaging,” ACS Chem. Biol., p. acschembio.8b00200, 2018. 
1056 
[71] 
W. P. Dempsey et al., “In vivo single-cell labeling by confined primed conversion,” Nat. Methods, vol. 12, 
1057 
no. 7, pp. 645–648, 2015. 
1058 
[72] 
M. A. Mohr and P. Pantazis, “Primed Conversion: The New Kid on the Block for Photoconversion,” 
1059 
Chem. – A Eur. J., vol. 24, no. 33, pp. 8268–8274, Jun. 2018. 
1060 
[73] 
B. Turkowyd et al., “A General Mechanism of Photoconversion of Green-to-Red Fluorescent Proteins 
1061 
Based on Blue and Infrared Light Reduces Phototoxicity in Live-Cell Single-Molecule Imaging,” Angew. 
1062 
Chemie - Int. Ed., vol. 56, no. 38, pp. 11634–11639, 2017. 
1063 
[74] 
X. Zhang et al., “Highly photostable, reversibly photoswitchable fluorescent protein with high contrast 
1064 
ratio for live-cell superresolution microscopy,” Proc. Natl. Acad. Sci., vol. 113, no. 37, pp. 10364–10369, 
1065 
Page 21 of 28
AUTHOR SUBMITTED MANUSCRIPT - JPhysD-122201.R1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Accepted Manuscript

----!@#$NewPage!@#$----
2016. 
1066 
[75] 
S. Wang et al., “GMars-T Enabling Multimodal Subdiffraction Structural and Functional Fluorescence 
1067 
Imaging in Live Cells,” Anal. Chem., vol. 90, pp. 6626–6634, 2018. 
1068 
[76] 
M. V Sednev, C. A. Wurm, V. N. Belov, and S. W. Hell, “Carborhodol: A New Hybrid Fluorophore 
1069 
Obtained by Combination of Fluorescein and Carbopyronine Dye Cores,” Bioconjug. Chem., vol. 24, no. 
1070 
4, pp. 690–700, Apr. 2013. 
1071 
[77] 
J. B. Grimm et al., “A general method to fine-tune fluorophores for live-cell and in vivo imaging,” Nat. 
1072 
Methods, vol. 14, p. 987, Sep. 2017. 
1073 
[78] 
P. Mateos-Gil, S. Letschert, S. Doose, and M. Sauer, “Super-Resolution Imaging of Plasma Membrane 
1074 
Proteins with Click Chemistry,” Front. Cell Dev. Biol., vol. 4, no. September, p. 98, 2016. 
1075 
[79] 
A. D. Thompson et al., “Long-Term Live-Cell STED Nanoscopy of Primary and Cultured Cells with the 
1076 
Plasma Membrane HIDE Probe DiI-SiR.,” Angew. Chem. Int. Ed. Engl., vol. 56, no. 35, pp. 10408–
1077 
10412, Aug. 2017. 
1078 
[80] 
G. Lukinavičius et al., “Fluorogenic probes for live-cell imaging of the cytoskeleton,” Nat. Methods, vol. 
1079 
11, no. 7, pp. 731–733, 2014. 
1080 
[81] 
E. Kozma and P. Kele, “Fluorogenic probes for super-resolution microscopy,” Org. Biomol. Chem., vol. 
1081 
17, no. 2, pp. 215–233, 2019. 
1082 
[82] 
S.-H. Shim et al., “Super-resolution fluorescence imaging of organelles in live cells with photoswitchable 
1083 
membrane probes,” Proc. Natl. Acad. Sci., vol. 109, no. 35, pp. 13978–13983, 2012. 
1084 
[83] 
M. Collot et al., “MemBright: A Family of Fluorescent Membrane Probes for Advanced Cellular Imaging 
1085 
and Neuroscience,” Cell Chem. Biol., vol. 26, no. 4, pp. 600-614.e7, 2019. 
1086 
[84] 
L. Wang, M. S. Frei, A. Salim, and K. Johnsson, “Small-Molecule Fluorescent Probes for Live-Cell Super-
1087 
Resolution Microscopy,” J. Am. Chem. Soc., vol. 141, no. 7, pp. 2770–2781, Feb. 2019. 
1088 
[85] 
L. D. Lavis, “Teaching Old Dyes New Tricks: Biological Probes Built from Fluoresceins and Rhodamines,” 
1089 
Annu. Rev. Biochem., vol. 86, no. 1, pp. 825–843, 2017. 
1090 
[86] 
A. Gautier et al., “An Engineered Protein Tag for Multiprotein Labeling in Living Cells,” Chem. Biol., vol. 
1091 
15, no. 2, pp. 128–136, Feb. 2008. 
1092 
[87] 
G. V Los et al., “HaloTag: a novel protein labeling technology for cell imaging and protein analysis.,” ACS 
1093 
Chem. Biol., vol. 3, no. 6, pp. 373–82, Jun. 2008. 
1094 
[88] 
B. A. Griffin, S. R. Adams, and R. Y. Tsien, “Specific covalent labeling of recombinant protein molecules 
1095 
inside live cells.,” Science (80-. )., vol. 281, no. 5374, pp. 269–72, Jul. 1998. 
1096 
[89] 
M. Lelek et al., “Superresolution imaging of HIV in infected cells with FlAsH-PALM.,” Proc. Natl. Acad. 
1097 
Sci. U. S. A., vol. 109, no. 22, pp. 8564–9, May 2012. 
1098 
[90] 
T. Stephan, A. Roesch, D. Riedel, and S. Jakobs, “Live-cell STED nanoscopy of mitochondrial cristae,” 
1099 
Page 22 of 28
AUTHOR SUBMITTED MANUSCRIPT - JPhysD-122201.R1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Accepted Manuscript

----!@#$NewPage!@#$----
Sci. Rep., vol. 9, no. 1, p. 12419, 2019. 
1100 
[91] 
S. N. Uno, D. K. Tiwari, M. Kamiya, Y. Arai, T. Nagai, and Y. Urano, “A guide to use photocontrollable 
1101 
fluorescent proteins and synthetic smart fluorophores for nanoscopy,” Microscopy, vol. 64, no. 4, pp. 
1102 
263–277, 2015. 
1103 
[92] 
H. Takakura et al., “Long time-lapse nanoscopy with spontaneously blinking membrane probes,” Nat. 
1104 
Biotechnol., vol. 35, no. 8, pp. 773–780, 2017. 
1105 
[93] 
C. Wang, M. Taki, Y. Sato, A. Fukazawa, T. Higashiyama, and S. Yamaguchi, “Super-Photostable 
1106 
Phosphole-Based Dye for Multiple-Acquisition Stimulated Emission Depletion Imaging,” J. Am. Chem. 
1107 
Soc., vol. 139, no. 30, pp. 10374–10381, 2017. 
1108 
[94] 
L. Chu et al., “Two-color nanoscopy of organelles for extended times with HIDE probes,” bioRxiv, 2019. 
1109 
[95] 
A. N. Butkevich et al., “Hydroxylated Fluorescent Dyes for Live-Cell Labeling: Synthesis, Spectra and 
1110 
Super-Resolution STED.,” Chemistry (Easton)., vol. 23, no. 50, pp. 12114–12119, Sep. 2017. 
1111 
[96] 
I. Nikić et al., “Debugging Eukaryotic Genetic Code Expansion for Site-Specific Click-PAINT Super-
1112 
Resolution Microscopy.,” Angew. Chem. Int. Ed. Engl., vol. 55, no. 52, pp. 16172–16176, Dec. 2016. 
1113 
[97] 
A. O. Khan, V. A. Simms, J. A. Pike, S. G. Thomas, and N. V Morgan, “CRISPR-Cas9 Mediated Labelling 
1114 
Allows for Single Molecule Imaging and Resolution.,” Sci. Rep., vol. 7, no. 1, p. 8450, Aug. 2017. 
1115 
[98] 
T. Schvartz et al., “Direct fluorescent-dye labeling of α-tubulin in mammalian cells for live cell and 
1116 
superresolution imaging.,” Mol. Biol. Cell, vol. 28, no. 21, pp. 2747–2756, Oct. 2017. 
1117 
[99] 
P. . B. T. . S. W. S. L. . W. R. . S. H. Wagner M.; Weber, “Light Dose is a Limiting Factor to Maintain Cell 
1118 
Viability in Fluorescence Microscopy and Single Molecule Detection,” Int. J. Mol. Sci., vol. 11, no. 3, pp. 
1119 
956–966, 2010. 
1120 
[100] 
M. E. Bulina, K. A. Lukyanov, O. V Britanova, D. Onichtchouk, S. Lukyanov, and D. M. Chudakov, 
1121 
“Chromophore-assisted light inactivation (CALI) using the phototoxic fluorescent protein KillerRed,” Nat. 
1122 
Protoc., vol. 1, no. 2, pp. 947–953, 2006. 
1123 
[101] 
M. M. Frigault, J. Lacoste, J. L. Swift, and C. M. Brown, “Live-cell microscopy - tips and tools,” J. Cell 
1124 
Sci., vol. 122, no. 6, pp. 753–767, 2009. 
1125 
[102] 
L. Song, E. J. Hennink, I. T. Young, and H. J. Tanke, “Photobleaching kinetics of fluorescein in 
1126 
quantitative fluorescence microscopy,” Biophys. J., vol. 68, no. 6, pp. 2588–2600, 1995. 
1127 
[103] 
J. W. Dobrucki, “Interaction of oxygen-sensitive luminescent probes Ru(phen)32+ and Ru(bipy)32+ with 
1128 
animal and plant cells in vitro: Mechanism of phototoxicity and conditions for non-invasive oxygen 
1129 
measurements,” J. Photochem. Photobiol. B Biol., vol. 65, no. 2, pp. 136–144, 2001. 
1130 
[104] 
J. A. Bloom and W. W. Webb, “Photodamage to intact erythrocyte membranes at high laser intensities: 
1131 
methods of assay and suppression.,” J. Histochem. Cytochem., vol. 32, no. 6, pp. 608–616, Jun. 1984. 
1132 
[105] 
C. M. Waterman-Storer, J. W. Sanger, and J. M. Sanger, “Dynamics of organelles in the mitotic spindles 
1133 
Page 23 of 28
AUTHOR SUBMITTED MANUSCRIPT - JPhysD-122201.R1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Accepted Manuscript

----!@#$NewPage!@#$----
of living cells: Membrane and microtubule interactions,” Cell Motil., vol. 26, no. 1, pp. 19–39, Jan. 1993. 
1134 
[106] 
W. Xiang, M. J. Roberti, J.-K. Hériché, S. Huet, S. Alexander, and J. Ellenberg, “Correlative live and 
1135 
super-resolution imaging reveals the dynamic structure of replication domains,” J. Cell Biol., vol. 217, no. 
1136 
6, pp. 1973–1984, Jun. 2018. 
1137 
[107] 
R. Kohen and A. Nyska, “Invited Review: Oxidation of Biological Systems: Oxidative Stress Phenomena, 
1138 
Antioxidants, Redox Reactions, and Methods for Their Quantification,” Toxicol. Pathol., vol. 30, no. 6, pp. 
1139 
620–650, Oct. 2002. 
1140 
[108] 
T.-W. Wu et al., “The cytoprotective effect of Trolox demonstrated with three types of human cells,” 
1141 
Biochem. Cell Biol., vol. 68, no. 10, pp. 1189–1194, Oct. 1990. 
1142 
[109] 
I. Rasnik, S. A. McKinney, and T. Ha, “Nonblinking and long-lasting single-molecule fluorescence 
1143 
imaging,” Nat. Methods, vol. 3, no. 11, pp. 891–893, 2006. 
1144 
[110] 
A. M. Bogdanov, E. I. Kudryavtseva, and K. A. Lukyanov, “Anti-Fading Media for Live Cell GFP Imaging,” 
1145 
PLoS One, vol. 7, no. 12, p. e53004, Dec. 2012. 
1146 
[111] 
P. Atkins and J. de Paula, Physical Chemistry, Eigth edit. New York, US: W.H. Freeman and Company, 
1147 
2006. 
1148 
[112] 
J. Widengren, A. Chmyrov, C. Eggeling, P.-Å. Löfdahl, and C. A. M. Seidel, “Strategies to Improve 
1149 
Photostabilities in Ultrasensitive Fluorescence Spectroscopy,” J. Phys. Chem. A, vol. 111, no. 3, pp. 429–
1150 
440, Jan. 2007. 
1151 
[113] 
H. Giloh and J. W. Sedat, “Fluorescence microscopy: reduced photobleaching of rhodamine and 
1152 
fluorescein protein conjugates by n-propyl gallate,” Science (80-. )., vol. 217, no. 4566, pp. 1252–1255, 
1153 
Sep. 1982. 
1154 
[114] 
K. Valnes and P. Brandtzaeg, “Retardation of immunofluorescence fading during microscopy.,” J. 
1155 
Histochem. Cytochem., vol. 33, no. 8, pp. 755–761, Aug. 1985. 
1156 
[115] 
G. D. Johnson, R. S. Davidson, K. C. McNamee, G. Russell, D. Goodwin, and E. J. Holborow, “Fading of 
1157 
immunofluorescence during microscopy: a study of the phenomenon and its remedy,” J. Immunol. 
1158 
Methods, vol. 55, no. 2, pp. 231–242, 1982. 
1159 
[116] 
A. Longin, C. Souchier, M. Ffrench, and P. A. Bryon, “Comparison of anti-fading agents used in 
1160 
fluorescence microscopy: image analysis and laser confocal microscopy study.,” J. Histochem. 
1161 
Cytochem., vol. 41, no. 12, pp. 1833–1840, Dec. 1993. 
1162 
[117] 
A. N. Fletcher and M. E. Pietrak, “Laser dye stability, part 10. Effects of DABCO on flashlamp pumping of 
1163 
coumarin dyes,” Appl. Phys. B, vol. 37, no. 3, pp. 151–157, 1985. 
1164 
[118] 
M. Ono, T. Murakami, A. Kudo, M. Isshiki, H. Sawada, and A. Segawa, “Quantitative Comparison of Anti-
1165 
Fading Mounting Media for Confocal Laser Scanning Microscopy,” J. Histochem. Cytochem., vol. 49, no. 
1166 
3, pp. 305–311, Mar. 2001. 
1167 
Page 24 of 28
AUTHOR SUBMITTED MANUSCRIPT - JPhysD-122201.R1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Accepted Manuscript

----!@#$NewPage!@#$----
[119] 
A. M. Bogdanov, E. A. Bogdanova, D. M. Chudakov, T. V Gorodnicheva, S. Lukyanov, and K. A. 
1168 
Lukyanov, “Cell culture medium affects GFP photostability: a solution,” Nat. Methods, vol. 6, no. 12, pp. 
1169 
859–860, 2009. 
1170 
[120] 
A. M. Edwards, E. Silva, B. Jofré, M. I. Becker, and A. E. De Ioannes, “Visible light effects on tumoral 
1171 
cells in a culture medium enriched with tryptophan and riboflavin,” J. Photochem. Photobiol. B Biol., vol. 
1172 
24, no. 3, pp. 179–186, 1994. 
1173 
[121] 
A. Grzelak, B. Rychlik, and G. Bartosz, “Light-dependent generation of reactive oxygen species in cell 
1174 
culture media,” Free Radic. Biol. Med., vol. 30, no. 12, pp. 1418–1425, 2001. 
1175 
[122] 
G. T. Spierenburg, F. T. J. J. Oerlemans, J. P. R. M. Van Laarhoven, and C. H. M. M. De Bruyn, 
1176 
“Phototoxicity of N-hydroxyethylpiperazine-N’-2-ethanesulfonic acid-buffered culture media for human 
1177 
leukemic cell lines,” Cancer Res., vol. 44, pp. 2253–2254, 1984. 
1178 
[123] 
J. S. Zigler, J. L. Lepe-Zuniga, B. Vistica, and I. Gery, “Analysis of the cytotoxic effects of light-exposed 
1179 
hepes-containing culture medium,” Vitr. Cell. Dev. Biol., vol. 21, no. 5, pp. 282–287, 1985. 
1180 
[124] 
P. Almada, S. Culley, and R. Henriques, “PALM and STORM: Into large fields and high-throughput 
1181 
microscopy with sCMOS detectors,” Methods, vol. 88, pp. 109–121, 2015. 
1182 
[125] 
S. Herbert, H. Soares, C. Zimmer, and R. Henriques, “Single-Molecule Localization Super-Resolution 
1183 
Microscopy: Deeper and Faster,” Microsc. Microanal., vol. 18, no. 6, pp. 1419–1429, 2012. 
1184 
[126] 
P. Almada et al., “Automating multimodal microscopy with NanoJ-Fluidics,” Nat. Commun., vol. 10, no. 1, 
1185 
p. 1223, 2019. 
1186 
[127] 
G. Donnert, C. Eggeling, and S. W. Hell, “Major signal increase in fluorescence microscopy through dark-
1187 
state relaxation,” Nat. Methods, vol. 4, no. 1, pp. 81–86, 2007. 
1188 
[128] 
K. König, T. W. Becker, P. Fischer, I. Riemann, and K.-J. Halbhuber, “Pulse-length dependence of 
1189 
cellular response to intense near-infrared laser pulses in multiphoton microscopes,” Opt. Lett., vol. 24, 
1190 
no. 2, pp. 113–115, 1999. 
1191 
[129] 
K. T. Takasaki, J. B. Ding, and B. L. Sabatini, “Live-Cell Superresolution Imaging by Pulsed STED Two-
1192 
Photon Excitation Microscopy,” Biophys. J., vol. 104, no. 4, pp. 770–777, 2013. 
1193 
[130] 
P. Bethge, R. Chéreau, E. Avignone, G. Marsicano, and U. V. Nägerl, “Two-Photon Excitation STED 
1194 
Microscopy in Two Colors in Acute Brain Slices,” Biophys. J., vol. 104, no. 4, pp. 778–785, 2013. 
1195 
[131] 
E. Schmidt and M. Oheim, “Two-photon imaging induces brain heating and calcium microdomain 
1196 
hyperactivity in cortical astrocytes,” bioRxiv, 2018. 
1197 
[132] 
B. Harke, J. Keller, C. K. Ullal, V. Westphal, A. Schönle, and S. W. Hell, “Resolution scaling in STED 
1198 
microscopy,” Opt. Express, vol. 16, no. 6, pp. 4154–4162, 2008. 
1199 
[133] 
G. Vicidomini et al., “Sharper low-power STED nanoscopy by time gating,” Nat. Methods, vol. 8, no. 7, 
1200 
pp. 571–573, 2011. 
1201 
Page 25 of 28
AUTHOR SUBMITTED MANUSCRIPT - JPhysD-122201.R1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Accepted Manuscript

----!@#$NewPage!@#$----
[134] 
Y. Wu and H. Shroff, “Faster, sharper, and deeper: structured illumination microscopy for biological 
1202 
imaging,” Nat. Methods, vol. 15, no. 12, pp. 1011–1019, 2018. 
1203 
[135] 
A. G. York et al., “Resolution doubling in live, multicellular organisms via multifocal structured illumination 
1204 
microscopy,” Nat. Methods, vol. 9, no. 7, pp. 749–754, 2012. 
1205 
[136] 
I. Gregor, M. Spiecker, R. Petrovsky, J. Großhans, R. Ros, and J. Enderlein, “Rapid nonlinear image 
1206 
scanning microscopy,” Nat. Methods, vol. 14, no. 11, pp. 1087–1089, 2017. 
1207 
[137] 
E. N. Ward and R. Pal, “Image scanning microscopy: an overview,” J. Microsc., vol. 266, no. 2, pp. 221–
1208 
228, 2017. 
1209 
[138] 
X. Huang et al., “Fast, long-term, super-resolution imaging with Hessian structured illumination 
1210 
microscopy,” Nat. Biotechnol., vol. 36, p. 451, Apr. 2018. 
1211 
[139] 
E. G. Reynaud, U. Kržič, K. Greger, and E. H. K. Stelzer, “Light sheet‐based fluorescence microscopy: 
1212 
More dimensions, more photons, and less photodamage,” HFSP J., vol. 2, no. 5, pp. 266–275, Oct. 2008. 
1213 
[140] 
E. G. Reynaud, J. Peychl, J. Huisken, and P. Tomancak, “Guide to light-sheet microscopy for 
1214 
adventurous biologists,” Nat. Methods, vol. 12, no. 1, pp. 30–34, 2014. 
1215 
[141] 
R. M. Power and J. Huisken, “A guide to light-sheet fluorescence microscopy for multiscale imaging,” 
1216 
Nat. Methods, vol. 14, no. 4, pp. 360–373, 2017. 
1217 
[142] 
F. Cella Zanacchi et al., “Live-cell 3D super-resolution imaging in thick biological samples,” Nat. Methods, 
1218 
vol. 8, no. 12, pp. 1047–1049, 2011. 
1219 
[143] 
F. Cella Zanacchi, Z. Lavagnino, M. Faretta, L. Furia, and A. Diaspro, “Light-Sheet Confined Super-
1220 
Resolution Using Two-Photon Photoactivation,” PLoS One, vol. 8, no. 7, p. e67667, Jul. 2013. 
1221 
[144] 
J. Kim et al., “Oblique-plane single-molecule localization microscopy for tissues and small intact animals,” 
1222 
Nat. Methods, vol. 16, no. 9, pp. 853–857, 2019. 
1223 
[145] 
P. Hoyer et al., “Breaking the diffraction limit of light-sheet fluorescence microscopy by RESOLFT,” Proc. 
1224 
Natl. Acad. Sci., vol. 113, no. 13, pp. 3442 LP – 3446, Mar. 2016. 
1225 
[146] 
Y. S. Hu et al., “Light-sheet Bayesian microscopy enables deep-cell super-resolution imaging of 
1226 
heterochromatin in live human embryonic stem cells,” Opt. Nanoscopy, vol. 2, no. 1, p. 7, 2013. 
1227 
[147] 
T. A. Planchon et al., “Rapid three-dimensional isotropic imaging of living cells using Bessel beam plane 
1228 
illumination,” Nat. Methods, vol. 8, no. 5, pp. 417–423, 2011. 
1229 
[148] 
L. Gao et al., “Noninvasive Imaging beyond the Diffraction Limit of 3D Dynamics in Thickly Fluorescent 
1230 
Specimens,” Cell, vol. 151, no. 6, pp. 1370–1385, 2012. 
1231 
[149] 
B.-J. Chang, V. D. Perez Meza, and E. H. K. Stelzer, “csiLSFM combines light-sheet fluorescence 
1232 
microscopy and coherent structured illumination for a lateral resolution below 100 nm,” Proc. Natl. Acad. 
1233 
Sci., vol. 114, no. 19, pp. 4869 LP – 4874, May 2017. 
1234 
[150] 
B.-C. Chen et al., “Lattice light-sheet microscopy: Imaging molecules to embryos at high spatiotemporal 
1235 
Page 26 of 28
AUTHOR SUBMITTED MANUSCRIPT - JPhysD-122201.R1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Accepted Manuscript

----!@#$NewPage!@#$----
resolution,” Science (80-. )., vol. 346, no. 6208, p. 1257998, Oct. 2014. 
1236 
[151] 
N. Chakrova, A. S. Canton, C. Danelon, S. Stallinga, and B. Rieger, “Adaptive illumination reduces 
1237 
photobleaching in structured illumination microscopy,” Biomed. Opt. Express, vol. 7, no. 10, pp. 4263–
1238 
4274, 2016. 
1239 
[152] 
J. Dreier et al., “Smart scanning for low-illumination and fast RESOLFT nanoscopy in vivo,” Nat. 
1240 
Commun., vol. 10, no. 1, p. 556, 2019. 
1241 
[153] 
T. Staudt, A. Engler, E. Rittweger, B. Harke, J. Engelhardt, and S. W. Hell, “Far-field optical nanoscopy 
1242 
with reduced number of state transition cycles,” Opt. Express, vol. 19, no. 6, pp. 5644–5657, 2011. 
1243 
[154] 
F. Göttfert et al., “Strong signal increase in STED fluorescence microscopy by imaging regions of 
1244 
subdiffraction extent,” Proc. Natl. Acad. Sci., vol. 114, no. 9, pp. 2125 LP – 2130, Feb. 2017. 
1245 
[155] 
J. Heine, M. Reuss, B. Harke, E. D’Este, S. J. Sahl, and S. W. Hell, “Adaptive-illumination STED 
1246 
nanoscopy,” Proc. Natl. Acad. Sci., vol. 114, no. 37, pp. 9797–9802, 2017. 
1247 
[156] 
D. Sage et al., “Super-resolution fight club: assessment of 2D and 3D single-molecule localization 
1248 
microscopy software,” Nat. Methods, vol. 16, no. 5, pp. 387–395, 2019. 
1249 
[157] 
A. Jimenez, K. Friedl, and C. Leterrier, “About samples, giving examples: Optimized Single Molecule 
1250 
Localization Microscopy,” Methods, 2019. 
1251 
[158] 
T. Dertinger, R. Colyer, G. Iyer, S. Weiss, and J. Enderlein, “Fast, background-free, 3D super-resolution 
1252 
optical fluctuation imaging (SOFI),” Proc. Natl. Acad. Sci., vol. 106, no. 52, pp. 22287 LP – 22292, Dec. 
1253 
2009. 
1254 
[159] 
S. Geissbuehler et al., “Live-cell multiplane three-dimensional super-resolution optical fluctuation 
1255 
imaging,” Nat. Commun., vol. 5, pp. 1–7, 2014. 
1256 
[160] 
S. Cox et al., “Bayesian localization microscopy reveals nanoscale podosome dynamics,” Nat. Methods, 
1257 
vol. 9, no. 2, pp. 195–200, 2012. 
1258 
[161] 
N. Gustafsson, S. Culley, G. Ashdown, D. M. Owen, P. M. Pereira, and R. Henriques, “Fast live-cell 
1259 
conventional fluorophore nanoscopy with ImageJ through super-resolution radial fluctuations,” Nat. 
1260 
Commun., vol. 7, no. 1, p. 12471, 2016. 
1261 
[162] 
R. F. Laine et al., “NanoJ: a high-performance open-source super-resolution microscopy toolbox,” J. 
1262 
Phys. D. Appl. Phys., vol. 52, no. 16, p. 163001, Feb. 2019. 
1263 
[163] 
S. Culley, K. L. Tosheva, P. Matos Pereira, and R. Henriques, “SRRF: Universal live-cell super-resolution 
1264 
microscopy,” Int. J. Biochem. Cell Biol., vol. 101, pp. 74–79, 2018. 
1265 
[164] 
R. J. Marsh et al., “Artifact-free high-density localization microscopy analysis,” Nat. Methods, vol. 15, no. 
1266 
9, pp. 689–692, 2018. 
1267 
[165] 
R. J. Marsh, S. Culley, and A. J. Bain, “Low power super resolution fluorescence microscopy by lifetime 
1268 
modification and image reconstruction,” Opt. Express, vol. 22, no. 10, pp. 12327–12338, 2014. 
1269 
Page 27 of 28
AUTHOR SUBMITTED MANUSCRIPT - JPhysD-122201.R1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Accepted Manuscript

----!@#$NewPage!@#$----
[166] 
L. von Chamier, R. F. Laine, and R. Henriques, “Artificial intelligence for microscopy: what you should 
1270 
know,” Biochem. Soc. Trans., vol. 47, no. 4, pp. 1029–1040, Jul. 2019. 
1271 
[167] 
C. Belthangady and L. A. Royer, “Applications, promises, and pitfalls of deep learning for fluorescence 
1272 
image reconstruction,” Nat. Methods, 2019. 
1273 
[168] 
M. Weigert et al., “Content-aware image restoration: pushing the limits of fluorescence microscopy,” Nat. 
1274 
Methods, vol. 15, no. 12, pp. 1090–1097, 2018. 
1275 
[169] 
W. Ouyang, A. Aristov, M. Lelek, X. Hao, and C. Zimmer, “Deep learning massively accelerates super-
1276 
resolution localization microscopy,” Nat. Biotechnol., vol. 36, p. 460, Apr. 2018. 
1277 
[170] 
L. Fang et al., “Deep Learning-Based Point-Scanning Super-Resolution Imaging,” bioRxiv, 2019. 
1278 
[171] 
S. Culley et al., “Quantitative mapping and minimization of super-resolution optical imaging artifacts,” Nat. 
1279 
Methods, vol. 15, p. 263, Feb. 2018. 
1280 
[172] 
G. Ball, J. Demmerle, R. Kaufmann, I. Davis, I. M. Dobbie, and L. Schermelleh, “SIMcheck: a Toolbox for 
1281 
Successful Super-resolution Structured Illumination Microscopy,” Sci. Rep., vol. 5, p. 15915, Nov. 2015. 
1282 
 
1283 
Page 28 of 28
AUTHOR SUBMITTED MANUSCRIPT - JPhysD-122201.R1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Accepted Manuscript