RESEARCH PAPER
Dep h- esol ed ibe pho ome y o amyloid
plaque signals in eely beha ing
Alzheime ’s disease mice
Nicole By on,aNiall McAlinden ,b,†Filippo Pisano ,c,d,e,†Ma co Pisanello ,
Jacques Fe ei a,aCinzia Mon ina o,cKei h Ma hieson,bMassimo De Vi o io,c,g
Fe uccio Pisanello ,cand Shuzo Saka a a,*
aUni e si y o S a hclyde, S a hclyde Ins i u e o Pha macy and Biomedical Sciences, Glasgow,
Uni ed Kingdom
bUni e si y o S a hclyde, SUPA, Ins i u e o Pho onics, Depa men o Physics, Glasgow, Uni ed Kingdom
cIs i u o I aliano di Tecnologia, Cen e o Biomolecula Nano echnologies, A nesano (Lecce), I aly
dUni e si y o Padua, Depa men o Physics and As onomy ‘G.Galilei’, Pado a, I aly
eUni e si y o Padua, Pado a Neu oscience Cen e , Pado a, I aly
Op ogeniX s. .l., A nesano (Lecce), I aly
gTechnical Uni e si y o Denma k, Depa men o Heal h Technology, Kongens Lyngby, Denma k
ABSTRACT. Signi icance: Cu en p eclinical e alua ion o Alzheime ’s disease pa hology in
mouse models elies on pos -mo em analyses, which hinde s he de elopmen and
op imiza ion o he apeu ic app oaches. Al hough in i o me hods exis , moni o ing
amyloid plaque signals ac oss mul iple b ain egions in eely beha ing animals
emains a signi ican challenge.
Aim: We aim o de elop an op ical app oach o add ess his challenge.
App oach: We used la and ape ed op ical ibe s in an Alzheime ’s mouse model.
Resul s: We i s con i med ha con en ional la ibe -based pho ome y can de ec
amyloid plaque signals ac oss mul iple b ain egions unde anes hesia a e injec ing
a blood-b ain-ba ie -pe meable ace , Me hoxy-X04. The dep h p o ile o in i o
luo escen signals is co ela ed wi h his ological signals. A machine lea ning
app oach could dis inguish be ween in i o luo escen signals o mice wi h and
wi hou amyloid plaques. Nex , a e alida ing he easibili y o dep h- esol ed ibe
pho ome y ex i o, we ch onically implan ed a ape ed ibe o moni o amyloid
plaque signals in eely beha ing mice. A e injec ing Me hoxy-X04, luo escen
signals inc eased in a dep h-speci ic manne in Alzheime ’s mice, bu no in hei
wild- ype li e ma es.
Conclusions: Ou app oach expands he capabili ies o ibe pho ome y o
moni o molecula pa hologies, such as amyloid plaques, e en in a eely beha ing
condi ion.
© The Au ho s. Published by SPIE unde a C ea i e Commons A ibu ion 4.0 In e na ional License.
Dis ibu ion o ep oduc ion o his wo k in whole o in pa equi es ull a ibu ion o he o iginal
publica ion, including i s DOI. [DOI: 10.1117/1.NPh.12.3.035014]
Keywo ds: Alzheime ’s disease; 5xFAD; ibe pho ome y; ape ed op ical ibe
Pape 25078GRR ecei ed Jun. 3, 2025; e ised Sep. 7, 2025; accep ed Sep. 9, 2025; published Sep.
23, 2025.
*Add ess all co espondence o Shuzo Saka a, [email p o ec ed]
†Equal con ibu ions.
Neu opho onics 035014-1 Jul–Sep 2025 •Vol. 12(3)
1 In oduc ion
Alzheime ’s disease (AD) is he mos common o m o demen ia. Amyloid plaques ha e long
been ecognized as a hallma k o AD.1–3Recen he apeu ics a ge ing amyloid-βp o o ib ils o
deposi ed amyloid plaques ha e p o en e ec i e in pa ien s4,5and success ul in ansla ing ea ly
p eclinical mouse da a6 o he clinic. As such, e alua ing no el in e en ions in p eclinical mouse
models plays a c i ical ole in accele a ing u he successes.
Al hough a wide ange o pha macological and non-pha macological in e en ion ap-
p oaches ha e been examined in mouse models, pos -mo em biochemical and his ological
analysis emains he gold s anda d in he ield. This means ha e e y in e en ion is a one- ime
expe imen ha equi es animal sac i ice. To op imize an in e en ion app oach lexibly, i would
be ideal o e alua e i s e icacy in eal- ime.
Al hough non-in asi e app oaches o assess a pa hological s a e a e a ailable o human
subjec s and plasma-based bioma ke de ec ion is an eme ging a ea,7–9op ions o mouse models
emain limi ed. Fo example, al hough mic odialysis allows assessmen o in e s i ial luid con-
en s in he b ain,10–12 i lacks spa ial esolu ion. Two-pho on imaging can de ec plaques in i o
by combining wi h a blood-b ain ba ie pe meable ace , Me hoxy-X04.13–15 Howe e , dep h
pene a ion is limi ed. I is c ucial o assess he plaque dis ibu ion a dep h, gi en he ac ha
amyloid pa hology can be seen ac oss mul iple deep b ain egions. Ano he app oach is op o-
acous ic omog aphy,16 allowing b ain-wide moni o ing o plaque signals. Howe e , cu en
echnologies do no pe mi moni o ing o amyloid pa hology ac oss mul iple b ain egions in
a eely beha ing condi ion.
Fibe pho ome y is an al e na i e, e sa ile op ical app oach.17–20 I allows moni o ing o
neu onal and non-neu onal popula ion ac i i y in a cell- ype-speci ic manne in i o.21–25
Al hough ibe pho ome y has also been adop ed o moni o in acellula signaling,26 i emains
unknown i ibe pho ome y can moni o ex acellula pa hological signals such as amyloid pla-
ques. Fo ins ance, ibe pho ome y ypically uses luo escen senso s ha equi e a ailabili y
and exp ession o hei gene ically encoded cons uc . To o e come his equi emen , molecula
pa hologies could be assessed h ough a non-gene ic s a egy by adminis a ion o a blood-b ain
ba ie luo escen plaque dye.
He e, we es he hypo hesis ha ibe pho ome y can be adap ed o access molecula pa h-
ologies, using a non-gene ic app oach. Speci ically, we es i i can be used o moni o amyloid
pa hology in 5xFAD mice, a widely used AD mouse model ca ying i e amilial AD mu a ions
ha de elop amyloid pa hology om 2 mon hs, which agg essi ely con inues wi h age.27 To his
end, we ake wo app oaches. Fi s , as a p oo -o -concep , we u ilize con en ional la op ical
ibe s o examine whe he amyloid pa hology can be moni o ed ac oss mul iple dep hs in mice.
Second, we exploi he pho onic p ope ies o ape ed op ical ibe s20 o es ablish dep h- esol ed
pho ome y o plaque signals ex i o and in i o. To moni o amyloid pa hology, we make use o
he blood-b ain-ba ie -pe meable luo escen compound, Me hoxy-x04. I s hyd ophobic s uc-
u e allows en y o he b ain, whe e i speci ically binds o be a-shee s ound wi hin amyloid
ib ils, allowing isualiza ion o amyloid plaques in Alzheime ’s disease models.13 Ou no el
pho ome y app oach expands he capabili ies o in i o ibe pho ome y o examine he pa ho-
logical ea u es o an AD mouse model in a eely beha ing condi ion.
2 Ma e ials and Me hods
2.1 Animals
Expe imen s we e pe o med in acco dance wi h he UK Animals (Scien i ic P ocedu es) Ac o
1986 Home O ice egula ions and app o ed by he Home O ice (PP0688944). Mice we e
housed wi h sex-ma ched li e ma es, i a ailable, on a 12-h/12-h ligh /da k cycle wi h access
o ood and wa e ad libi um. All expe imen s we e pe o med du ing he ligh pe iod. 5xFAD
mice27 (JAX006554, The Jackson Labo a o y, Ba Ha bo , Maine, Uni ed S a es) we e ob ained
and backc ossed on o C57BL/6 backg ound (>F10). Male and emale 5xFAD+/−, e e ed o
as 5xFAD, and 5xFAD−/−, e e ed o as WT, mice aged 3 o 12 mon hs we e used (Table S1 in
he Supplemen a y Ma e ial). Fi een mice (se en 5xFAD, six males and one emale; eigh WT,
i e males and h ee emales) we e used o in i o la ibe pho ome y expe imen s. Th ee mice
By on e al.: Dep h- esol ed ibe pho ome y o amyloid plaque signals. . .
Neu opho onics 035014-2 Jul–Sep 2025 •Vol. 12(3)
( wo 5xFAD, wo males; one male WT) we e used o ex i o ape ed ibe pho ome y expe i-
men s. Twen y- i e mice (14 5xFAD, 6 males and 8 emales; 11 WT, 4 males and 7 emales) we e
used o in i o ape ed op ical ibe (TF)-based pho ome y expe imen s. Some mice we e
excluded om he analysis due o eco ding o his ological issues, including an incomple e
eco ding, inaccu a e lase powe es ima ions, o inaccu a e egis a ion o his ological sec ions
(Table S1 in he Supplemen a y Ma e ial). No blinding o andomiza ion was adop ed, no was
he e an analysis o sex di e ences, as his was a echnical de elopmen s udy.
2.2 In Vi o Fla Fibe Pho ome y
2.2.1 Pho ome y sys em
A la ibe -based pho ome y sys em is desc ibed elsewhe e.23 To adjus he sys em o
Me hoxy-X04 signals, he ligh om a 405-nm ligh -emi ing diode (LED) (M405L3, Tho labs,
New on, New Je sey, Uni ed S a es) was collima ed by an asphe ic lens (AL2520M-A,
Tho labs), passed a bandpass il e (FB405-10, Tho labs), e lec ed o wo dich oic mi o s
(MD498 and DMLP425R, Tho labs), and ocused by an asphe ic lens on o he mul imode pa ch
cable (400-μmco e, 0.5 NA; MAF3L1, Tho labs) and la ibe (200-μmco e, 0.50 NA;
MAF3L1, Tho labs). Emi ed ligh passed back h ough he ibe and pa ch cable and was colli-
ma ed by an asphe ic lens, passed a dich oic mi o (DMLP425R, Tho labs), and was edi ec ed
owa d an emission il e by a b oadband mi o (BB1-E02, Tho labs). Fil e ed ligh was ocused
by an asphe ic lens on o a pho ode ec o (NewFocus 2151, Newpo ) o measu emen . In o al,
440- and 550-nm emission il e s (FB440-10 and FB550-10, Tho labs) we e used in e change-
ably by adding and emo ing a d op-in il e holde (DCP1, Tho labs) be ween each measu e-
men . A NIDAQ de ice (NI USB-6211, Na ional Ins umen s, Aus in, Texas, Uni ed S a es) and
cus om LabVIEW so wa e we e used o con ol he LED and pho ode ec o .
2.2.2 Pho ome y expe imen s
5xFAD and WT mice we e used o in i o la ibe dep h p o ile pho ome y expe imen s.
A o al o 10 mg∕kg o Me hoxy-X04 (4920, Toc is, B is ol, Uni ed Kingdom) dissol ed in
45% p opylene glycol/45% 0.1 M phospha e-bu e ed saline (PBS)/10% DMSO was adminis-
e ed in ape i oneally. The nex day, mice we e anes he ized wi h u e hane (1.5 g∕kg) and
placed in a s e eo axic ame (Model 963, KOPF Ins umen s, Los Angeles, Cali o nia, Uni ed
S a es). Body empe a u e was main ained a 37°C (ATC 100, Wo ld P ecision Ins umen s,
Sa aso a, Flo ida, Uni ed S a es), and eye gel (Hylonigh o Visco ea s) was applied h oughou .
C anio omies we e made o e h ee si es (si e 1: AP þ0.49 mm; ML 0.25 mm; si e 2: AP
−1.79 mm, ML 1.50 mm; si e 3: AP −3.30 mm, ML 2.80 mm), which we e co e ed wi h
KWIK-SIL (Wo ld P ecision Ins umen s). Th oughou hese p ocedu es, he anes he ic le el
was main ained wi h iso lu ane (1% o 1.5% a 0.8 L∕min ai low).
Mice we e mo ed o he eco ding se -up (SR-8N-S, Na ishige, Tokyo, Japan) o dep h
p o ile measu emen s, whe e body empe a u e was main ained a 37°C (50-7212, Ha a d
Appa a us, Hollis on, Massachuse s, Uni ed S a es). Fo each c anio omy, KWIK-SIL was
emo ed, and he la ibe (200-μmco e, 0.50 NA; MAF3L1, Tho labs) was lowe ed o he
b ain su ace by he subs age and mic omanipula o (SM-15M and SM-15, Na ishige). A mea-
su emen was comple ed wi h bo h 440- and 550-nm emission il e s, be o e epea ing his
p ocess a inc easing 100-μmdep hs un il a maximum dep h o 4000 μmwas eached.
Fo each dep h measu emen , he LED wen h ough 10 epe i ions o 10 ms on, 5 ms o ,
sepa a ed by a 10-ms baseline o ligh powe s anging om 0.1 o 1mW∕mm2. Da a we e
collec ed a he pho ode ec o and LED sync channel a 5000 Hz.
2.2.3 His ological analysis
Immedia ely a e pho ome y expe imen s, mice we e deeply anes he ized wi h lidocaine (2%)
and pen oba bi al (200 mg∕ml) and ansca dially pe used wi h PBS and 4% pa a o maldehyde.
B ain issue was emo ed and pos - ixa ed in he same ixa i e a 4°C o e nigh , and c yop o-
ec ed in 30% suc ose/0.1 M PBS o se e al days. Then, 100-μmco onal sec ions we e p epa ed
using a sliding mic o ome (SM2010R, Leica, We zla , Ge many). As Me hoxy-X04 was injec ed
he day be o e b ain e ie al, amyloid pa hology was s ained acco dingly, and no o he s aining
By on e al.: Dep h- esol ed ibe pho ome y o amyloid plaque signals. . .
Neu opho onics 035014-3 Jul–Sep 2025 •Vol. 12(3)
was necessa y. The e o e, sec ions we e washed in PBS (3×5 min), be o e insing in gela in
and moun ing on o glass slides. Once d y, slides we e sealed wi h co e slips and luo omoun
solu ion (The moFishe Scien i ic Fluo omoun G, In i ogen, Wal ham, Massachuse s,
Uni ed S a es). The b ain sec ions an e io and pos e io o and con aining he la ibe implan
si e we e imaged. Images we e acqui ed using an up igh luo escence mic oscope (Eclipse
E600, Nikon, Tokyo, Japan) and CMOS came a (C11440-36U, Hamama su, Shizuoka, Japan)
a 4×magni ica ion.
The pos -p ocessing o his ological images is summa ized in Fig. S1 in he Supplemen a y
Ma e ial. Images we e s i ched au oma ically, and he s i ched images we e egis e ed o he
Allen Mouse B ain Common Coo dina e F amewo k using a modi ied e sion o a eely a ail-
able so wa e (AMaSiNe, h ps://gi hub.com/ snnlab/AMaSiNe).28 Plaques we e au oma ically
de ec ed as pa icles be ween 11 and 20 μmin diame e . Coo dina es o he la ibe ack we e
manually anno a ed ac oss sec ions and es ima ed using a linea eg ession model.
To es ima e he plaque densi y, as he ipsila e al hemisphe e was damaged by he la ibe ,
he es ima ed ibe ack was p ojec ed on o he con ala e al hemisphe e, assuming ha plaques
a e dis ibu ed homogeneously ac oss hemisphe es. To quan i y plaques, a cylinde wi h a adius
o 250 μmwas gene a ed along he p ojec ed ibe ack, and plaques wi hin 250 μm om each
dep h measu e we e coun ed e e y 100-μms ep. As i ually no plaque signals we e de ec ed in
WT mice, andomly gene a ed noise anging be ween 0 and 1×10−10 was added o he his o-
logical quan i ica ion o bo h 5xFAD and WT mice wi hou p oducing a i icial ends. Plaque
densi y was de e mined by calcula ing he plaque coun pe sphe ical olume a each quan i-
ied dep h.
2.2.4 Pho ome y signal p ocessing
O line analysis was pe o med using cus om MATLAB codes. Fo all analyses, 440-nm pho-
ome y da a a 1mW∕mm2we e used. To ob ain a dep h p o ile, z-sco ed luo escence a each
dep h was calcula ed by dF ¼F−meanðF0Þ
s dðF0Þ, whe e Fis he aw luo escence and F0is he mean
luo escence ac oss 0 o 400 μm. A median il e was applied o smoo h da a (window size: 4).
Fo compa ison o pho ome y o his ology signals, Spea man’s ho co ela ion es was
comple ed.
To classi y pho ome y signals, each dep h p o ile was z-sco ed, and he da a ma ix
con aining all samples (animals and eco ding si es) was cons uc ed. As signals we e sampled
a 41 dep hs o each pene a ion, p incipal componen analysis was applied o educe he dimen-
sion. The i s h ee p incipal componen s (PCs) we e used o ain a suppo ec o machine
model o p edic geno ype. The aining was done by aking he lea e-one-ou p ocedu e, and
he geno ype o he emaining sample was p edic ed by he ained model. The same p ocedu es
we e epea ed ac oss all samples. To assess he pe o mance o his classi ica ion, a con usion
ma ix was compu ed.
2.3 Ex Vi o Tape ed Fibe Pho ome y
2.3.1 Pho ome y sys em
A cus om-buil wo-pho on lase scanning mic oscope desc ibed in de ail in Re . 29 was used
o e alua e he capabili y o TFs o ga he signi ican da a om plaque-o igina ed luo escence.
The mic oscope is equipped wi h h ee di e en imaging channels: an epi- luo escence channel
p o ided wi h a dich oic mi o (FF665-Di02, Sem ock, Wes Hen ie a, New Yo k, Uni ed
S a es) and bandpass il e (FF01-520/70, Sem ock) o acqui e he e e ence image o he b ain
slices and he TF; a ibe -coupled channel p o ided wi h a bandpass il e (FF01-442/42-25,
Sem ock) and synch onized o he mic oscope scanne o acqui e he TF ligh collec ion ield;
and a wide- ield channel p o ided wi h a bandpass il e (FBH450-40, Tho labs) o acqui e he
TF illumina ion ield on he senso o a sCMOS came a (O ca Flash li e 4.0, Hamama su).
2.3.2 Ex i o pho ome y expe imen s
To label amyloid pa hology, Me hoxy-X04 (10 mg∕kg) (4920, Toc is) was injec ed in ape i o-
neally. The ollowing day, b ain samples we e e ie ed as desc ibed abo e. Howe e , o
By on e al.: Dep h- esol ed ibe pho ome y o amyloid plaque signals. . .
Neu opho onics 035014-4 Jul–Sep 2025 •Vol. 12(3)
his p ocedu e, he sample did no unde go c yop o ec ion in suc ose and was a he s o ed in
PBS. Then, 500-μmco onal sec ions we e p epa ed using a ib a ome (054018, Ted Pella
Inco po a ed, Redding, Cali o nia, Uni ed S a es) and s o ed in PBS.
A b ain slice con aining he sep al a ea was placed in he imaging plane o he cus om-buil
wo-pho on lase scanning mic oscope, and a TF (0.39 NA, 200-μmco e, 225-μmcladding,
1.8-mm emission leng h, Op ogeniX, A nesano, I aly) was inse ed in o he slice so ha he
1.8-mm op ically ac i e egion o he ape was ully wi hin he sep al a ea.
Fem osecond pulsed lase ligh a 740 nm was used o elici Me hoxy-X04 luo escence in
b ain slices h ough he mic oscope o measu e he e e ence image and TF collec ion.
Con inuous wa e 405-nm lase ligh (MDL-III-405-50-mW, CNI lase ) was used o elici
Me hoxy-X04 luo escence in b ain slices h ough he TF o measu e he TF illumina ion ield.
Con ol on he angle a which ligh couples o he ibe was achie ed using a gal o mi o -based
scanning sys em iden ical o he one employed o he in i o expe imen s (desc ibed la e in
he ex ).
2.3.3 Ex i o da a analysis
Images we e p ocessed wi h he FIJI so wa e30 and cus om Py hon sc ip s. Images acqui ed in
he e e ence and ibe channels we e in insically egis e ed by syncing bo h pho ode ec o s
(PMTs) wi h he lase scanning. Illumina ion ields acqui ed on a cha ge-coupled de ice
(CCD) came a we e escaled and manually egis e ed on he e e ence and ibe image. All
images we e p ep ocessed wi h backg ound sub ac ion and h esholding be o e pe o ming
au oma ed amyloid plaque localiza ion and coun ing (analyze pa icles). The dis ibu ion o he
plaques along he ibe axis was calcula ed on he images om he e e ence and ibe channels
by compu ing he numbe o plaque cen e s alling wi hin a ci cula egion o in e es (ROI) o
adius 250 μmand cen e ed on he ibe axis ha was mo ed along he ibe a s eps o 40 μm.
The co ela ion be ween he quan i ied plaques in he e e ence image and ibe collec ion image
was calcula ed using Spea man’s ho co ela ion es .
The image s ack o he illumina ed a eas o issue agains he inpu angles (gal o ol ages)
was p e-p ocessed by backg ound sub ac ion and no malized o he maximum alue o he s ack,
o accoun o he dynamic in illumina ion densi y. Fo each ame—co esponding o an inpu
angle—a cen oid o he illumina ion peak was calcula ed a e sub ac ing an in ensi y baseline
co esponding o he in ensi y o he e minal inpu angles, whe e a negligible amoun o ligh is
deli e ed in he issue.
The pho ome y s ack was hen calcula ed by mul iplying he illumina ion s ack agains he
ibe channel image pixel by pixel, as desc ibed in Re . 29. The simula ed pho ome y in ensi ies
we e calcula ed om he in eg al in ensi y in he pho ome y s ack. Each alue o pho ome y
in ensi y was hen a ibu ed o a spa ial loca ion along he ibe co esponding o he cen oid o
he peak o he illumina ion p o ile o he co esponding inpu angle. This allowed us o compa e
he pho ome y in ensi y p o iles agains he plaque densi y p o ile in he e e ence image (in en-
si y scaled by di iding by he maximum alue). The co ela ion be ween he pho ome y in ensi y
p o iles and he in ensi y-scaled plaque densi y in he e e ence image was calcula ed using
Spea man’s ho co ela ion es .
2.4 In Vi o Tape ed Fibe Pho ome y
2.4.1 Pho ome y sys em
A 405-nm lase (MDL-III-405-50-mW, CI90055, CNI lase s o Cobol 06-01 Se ies, Hubne
Pho onics, Solna, Sweden) deli e ed exci a ion ligh . A po ion o his ligh beam was e lec ed
owa d a pho ode ec o (PDA25K-EC, Tho labs) by a glass slide o moni o lase s abili y.
The es passed b oadband mi o s be o e being ocused on o a gal o mi o (GVS001, Tho labs)
by an asphe ic lens (AC254-050-A-ML, Tho labs). The gal o mi o con olled he angle o
e lec ed ligh , and he e o e, he angle o ligh coupled in o he ape ed ibe (0.39 NA, 200-μm
co e, 225-μmcladding, 1.8-mm emission leng h, Op ogeniX). Re lec ed ligh was collima ed by
an asphe ic lens (AC254-050-A-ML, Tho labs), passed h ough a dich oic mi o (DMSP425T,
Tho labs), and hen ocused by an asphe ic lens (AC254-030-A-ML, Tho labs) on o a mul imode
pa ch cable (200-μmco e, 0.39° NA) and an implan ed ape ed ibe . Emi ed ligh passed back
By on e al.: Dep h- esol ed ibe pho ome y o amyloid plaque signals. . .
Neu opho onics 035014-5 Jul–Sep 2025 •Vol. 12(3)
h ough he ape ed ibe and pa ch cable and was collima ed by an asphe ic lens and e lec ed o
a dich oic mi o (DMSP425T, Tho labs) owa d a dich oic mi o (DMSP490R, Tho labs),
which sepa a ed 440- and 550-nm ligh beams, wi h he use o emission il e s (FB440-10 and
FB550-10, Tho labs). Each beam was hen ocused on o he pho ode ec o by an asphe ic lens
(AC254-030-A-ML, Tho labs) o measu emen . A NIDAQ de ice (NI USB-6343, Na ional
Ins umen s) and cus om LabVIEW so wa e we e used o con ol he lase , gal o mi o , and
pho ode ec o s.
2.4.2 Calib a ion and pho ome y p o ocols
Fo equalizing he ligh powe ac oss he ape ed ibe , he ibe was placed in a uni o m
Me hoxy-X04 solu ion (0.1 mM) and unde wen an illumina ion p o ocol. The lase was ac i-
a ed in cycles o 10 ms on and 5 ms o a 60 and 80 μW o i e epe i ions. In pa allel wi h lase
on pe iods, he gal o mi o was d i en om −1 o 4.5 V, emaining a 0 V du ing he o pe iod.
Da a we e collec ed a 5000 Hz. The luo escence a each gal o measu e was used o calcula e
he equi ed powe o uni o m luo escence ac oss he ape ed ibe . Fo each gal o measu e,
he mean luo escence was ans o med o a ela i e equi ed powe by PR¼1
F÷minð1
FÞ, whe e
PRis he equi ed powe and Fis he luo escence.
2.4.3 Pho ome y expe imen s
5xFAD and WT mice we e used. Mice we e anaes he ized wi h iso lu ane (5% o induc ion and
1% o 1.5% o main enance wi h 0.8 L∕min ai low). Mice we e placed in a s e eo axic ame
(Model 963, KOPF Ins umen s), e lexes we e moni o ed, body empe a u e was main ained a
37°C (ATC 100, Wo ld P ecision Ins umen s), and eye gel (Hylonigh o Visco ea s) was
applied h oughou . Na opin (8mg∕kg) was adminis e ed o he su gical a ea subcu aneously,
while Ve e gesic (0.1 mg∕kg), Rimadyl (20 mg∕kg), and saline (0.3 ml, FKE1323, Bax e ,
Dee ield, Illinois, Uni ed S a es) we e adminis e ed subcu aneously o he ump. Skull sc ews
(418 o 7123, RS Componen s, Co by, Uni ed Kingdom) we e implan ed in he skull: wo in
he on (AP þ1.50 mm,ML1.00 mm), one o e he igh hemisphe e (AP −1.00 mm,ML
þ4mm), one o e he le hemisphe e (AP −3.00 mm,ML−4.00 mm), and wo in he back
(AP −2.00 mm om lambda, ML 2.00 mm) as ancho s. The ape ed ibe was implan ed a he
a ge si e (AP −3.50 o −3.60 mm, ML 2.25 o 2.80 mm, DV 2.50 o 2.00 mm) be o e being
secu ed wi h KWIK-KAST (Wo ld P ecision Ins umen s) and laye s o supe glue (918-6872,
RS-P o) and den al cemen (Kemden ). Den al cemen was applied ac oss he skull and skin
edges. A e 5 days o eco e y, mice we e habi ua ed o handling and being e he ed o he
pho ome y sys em. Al hough sc u ed, hey we e a ached o he pa ch cable ia a ma ing slee e
(ADAL1-5, Tho labs) and placed in o a eco ding chambe (46.5 ×21.5 ×20.5 cm). Reco dings
included a baseline measu emen wi h no Me hoxy-X04 o 5 h. Twen y- ou hou s la e , a 5-h
measu emen was comple ed, wi h Me hoxy-X04 (10 mg∕kg) (4920, Toc is) being injec ed
in ape i oneally a 30 min. Twen y- ou hou s la e , he luo escence was moni o ed o a u he
2 h. The lase was ac i a ed in cycles o 10 ms on, 5 ms o ac oss ou powe s anging om 60
o 140 μW. In pa allel wi h lase on pe iods, he gal o mi o was d i en om −1 o 4.5 V,
emaining a 0 V du ing he OFF pe iod. This p o ocol was epea ed h ee o i e imes a
a sampling in e al o 60 o 300 s. Da a we e collec ed a 5000 Hz. A e se e al eco dings,
mice we e culled o his ological analysis.
2.4.4 His ological analysis
The b ain sec ions (50 μm) we e p epa ed as desc ibed abo e. As Me hoxy-X04 was injec ed
se e al days be o e b ain e ie al, amyloid pa hology will be s ained acco dingly, bu sec ions
we e co-s ained wi h ano he plaque ma ke as a backup. To iden i y he TF ack, sec ions
we e co-s ained wi h a mic oglial ma ke . A e washing in PBS (3×5 min), he sec ions we e
incuba ed in 10% no mal goa se um in 0.3% T i on X in PBS (PBST) o 1 h. Then, he sec-
ions unde wen p ima y an ibody incuba ion wi h an i-Iba1 an ibody (1:1000 in 3% no mal goa
se um in PBST; Abcam, ca . no. ab178846) o e nigh a 4°C. A e washes in PBS (3×5 mins),
a 2-h seconda y an ibody incuba ion wi h Alexa Fluo 594 (1:1000 in 3% no mal goa se um
in PBST; The moFishe , ca . no. A-11005) was comple ed. Sec ions we e washed in PBS
By on e al.: Dep h- esol ed ibe pho ome y o amyloid plaque signals. . .
Neu opho onics 035014-6 Jul–Sep 2025 •Vol. 12(3)
(3×5 mins) be o e co-s aining wi h Thio la in-S (0.01% in PBS; Sigma-Ald ich, ca . no.
T1892, S . Louis, Missou i, Uni ed S a es) o 15 min. The sec ions we e washed in PBS
(3×5 min) be o e insing in gela in and moun ing on o glass slides. Once d y, slides we e sealed
wi h co e slips and luo omoun solu ion (The moFishe Scien i ic Fluo omoun G, In i ogen).
The b ain sec ions we e imaged as desc ibed abo e.
Image p ocessing was he same as desc ibed abo e. Plaque quan i ica ion was he same as
desc ibed abo e bu was comple ed a 41 s eps e e y 39.5 μm(gal o measu e esolu ion). Also,
o accoun o o e es ima ions occu ing du ing he es ima ion o he TF pene a ion ack,
we shi ed he pene a ion ack up 600 μm[Fig. S2(a) in he Supplemen a y Ma e ial].
2.4.5 Pho ome y signal p ocessing
O line analysis was pe o med using cus om MATLAB codes. To es ima e Me hoxy-X04-
ela ed luo escen signals, he dynamics o he au o luo escence we e p edic ed based on he
signals wi hou Me hoxy-X04 adminis a ion (day 0). To his end, a linea model was es ablished
a each gal o scanning le el as AF ¼að Þþb, whe e AF was p edic ed au o luo escence, was
he ime samples, and aand ba e he model coe icien s, which we e es ima ed by wo old c oss-
alida ion. Me hoxy-X04 signals we e es ima ed a each gal o scanning le el as Fm¼F−AF,
whe e Fmwas he es ima ed Me hoxy-X04 signals and Fwas aw luo escen signals.
Measu emen s om 0- o 4.5-V gal o scanning le els we e aken in 5-min ime bins ac oss
eco dings. The dep hs con aining he minimum and maximum median signal om 30 o 240 min
we e z-sco ed by dF ¼Fm−meanðF0Þ
s dðF0Þ, whe e dF was he z-sco ed luo escence, Fmwas es ima ed
om he Me hoxy-X04 signals, and F0was he mean luo escence in he i s 25 min o eco d-
ing. Da a we e smoo hed using a mo ing median il e . Fo all analyses, 440-nm pho ome y da a
a 120 μWwe e used.
2.5 S a is ical Analysis
All s a is ical analyses we e pe o med using MATLAB unless o he wise s a ed. A Kolmogo o –
Smi no es was comple ed o de e mine he no mali y o he da a. Acco dingly, in Figs. 1–3,
co ela ion analysis was comple ed using Spea man’s ho es . A wo-sample - es was pe -
o med in Figs. 1and 3 o compa e he co ela ion coe icien s. In Fig. 3, he maximum and
minimum luo escen p o iles o 5xFAD and WT we e es ed wi h a Wilcoxon signed- ank es .
All da a we e shown as mean ± s anda d e o o mean unless o he wise s a ed.
3 Resul s
3.1 Fla Fibe -Based Pho ome y o Moni o Amyloid Pa hology
in Anes he ized 5xFAD Mice
As a p oo -o -concep expe imen , we i s examined whe he con en ional la ibe -based pho-
ome y [Fig. 1(a)] allows moni o ing o plaque signals ac oss mul iple dep hs in i o. To his
end, Me hoxy-X04 was injec ed a day be o e expe imen s in ei he 5xFAD mice (n¼6) o hei
li e ma e wild- ype (WT) con ols (n¼7). Unde u e hane anes hesia, we pe o med ibe pho-
ome y a inc easing dep hs by g adually lowe ing a la ibe in o he b ain [Fig. 1(b)]. As we
lowe ed he la ibe , we obse ed a iable luo escen signals along he pene a ion ack in a
5xFAD mouse [Fig. 1(c)]. A e pho ome y expe imen s, we pe o med his ological analysis o
econs uc he dep h p o ile o plaque densi y along he pene a ion ack [Fig. 1(c)]. We ook
he dep h p o ile om he con ala e al hemisphe e o es ima e signals in he in ac b ain based
on p e ious li e a u e ha demons a es homogenous labeling o amyloid pa hology ac oss
b ain egions and hemisphe es.31 We con i med a signi ican posi i e co ela ion be ween in i o
luo escen and pos -mo em his ological dep h p o iles ( ¼0.855,p<0.0001, Spea man’s
ho) [Fig. 1(c)]. The dep h-dependen end o lowe co ela ion is likely a ibu able o ana omi-
cal ac o s ha a y ac oss egions and mice. In con as , signals in a con ol animal showed
lowe a iabili y. As he con ol animal did no o m plaques, no co ela ion was ound [Fig.
S3(a) in he Supplemen a y Ma e ial]. We epea ed he same expe imen s ac oss h ee di e en
si es ac oss animals o ind s a is ically signi ican posi i e co ela ions in 5xFAD mice
(0.499 0.081) compa ed wi h con ol mice (−0.101 0.061)( ð22Þ¼5.70,p<0.0001,
By on e al.: Dep h- esol ed ibe pho ome y o amyloid plaque signals. . .
Neu opho onics 035014-7 Jul–Sep 2025 •Vol. 12(3)
Fig. 1 Fla ibe -based pho ome y ealizes luo escen signals e lec i e o amyloid pa hology in
5xFAD mice. (a) Fla ibe -based pho ome y sys em allows 405-nm LED ligh o pass h ough he
la ibe implan ed in ei he 5xFAD o WT mouse b ain. Emi ed luo escence ligh passes back
h ough he la ibe be o e being il e ed o cu ou exci a ion ligh and being ocused on o a PMT.
The ligh pa h was con olled by a se ies o mi o s (M), il e s (F), and lenses (L). (b) Schema ic o
he expe imen al design. Mice we e injec ed wi h Me hoxy-X04 he day be o e being e minally
anes he ized o pho ome y eco dings. A e , b ain issue was e ie ed o his ology, whe e
pho ome y and his ology signals unde wen co ela ion analysis. Pa o he diag am was c ea ed
in BioRende . (c) Le : example in i o luo escen and pos -mo em his ology dep h p o iles. Solid
By on e al.: Dep h- esol ed ibe pho ome y o amyloid plaque signals. . .
Neu opho onics 035014-8 Jul–Sep 2025 •Vol. 12(3)
wo-sample - es ) [Fig. 1(d)]. Ou indings also con i m posi i e co ela ion in he implan ed
(ipsila e al) hemisphe e, wi h compa able plaque densi ies ac oss hemisphe es (Fig. S4 in he
Supplemen a y Ma e ial). These esul s indica e ha in i o pho ome y signals e lec plaque
signals a dep h.
We also examined whe he pho ome y-based dep h p o iles alone can disc imina e geno-
ypes (i.e., 5xFAD o WT). To his end, we ook a machine lea ning app oach. A e educing
he dimension o 3 om 41 da a poin s ac oss he pene a ion by pe o ming p incipal compo-
nen analysis, we ained a suppo ec o machine model by lea ing one sample ou . Then, we
es ed he model o assess i he emaining sample was accu a ely classi ied [Fig. 1(e)]. The
o e all pe o mance was 83.3%: ou o 13 5xFAD eco dings, 10 we e p ope ly classi ied
(76.9% hi ), whe eas ou o 11, 10 WT eco dings we e co ec ly ejec ed (90.9%). These esul s
indica e ha in i o pho ome y signals alone a e capable o classi ying geno ypes.
3.2 Ex Vi o E alua ion o Tape ed Fibe -Based Pho ome y
Gi en he success o he la ibe -based app oach, we sough o de elop a me hod o moni o
amyloid pa hology ac oss b ain egions in eely beha ing animals. To his end, we adop ed TF
echnology.20,32 Be o e using TFs in i o, we examined i a TF allowed dep h- esol ed pho om-
e y o amyloid plaques ex i o in a con olled, ele an en i onmen (Fig. 2). To his end,
we injec ed Me hoxy-X04 in o a 5xFAD mouse and ex ac ed he b ain issue on he ollowing
day. We hen inse ed a TF (0.39NA, 200-μmco e, 225-μmcladding, 1.8-mm emission leng h)
in a 500-μm- hick co onal b ain slice con aining he sep al a ea and used a wo-pho on scanning
mic oscope equipped wi h an epi- luo escence channel and a ibe -coupled channel [Fig. 2(a)]
29,33 o simul aneously collec a e e ence image o he b ain slice [Fig. 2(c)] and a measu emen
o he TF collec ion ield Cðx; yÞ, whe e ðx; yÞdeno es a loca ion in he slice [Fig. 2(d)]. A simila
app oach was desc ibed in p e ious wo ks.20,29 To de e mine i he ibe collec ion ield was
la ge enough o ga he a signi ican amoun o plaque-o igina ed luo escence, we compa ed
he numbe o plaques enclosed inside a 250-μm adius sphe e wi h he cen e mo ing along
he TF axis on hese wo images [Fig. 2(g)]; despi e he ibe collec ing a lowe numbe o pla-
ques han he epi- luo escence channel, mainly due o he decay in collec ion e iciency while
mo ing adially a he om he ibe axis, he wo cu es a e consis en ( ¼0.884,p<0.0001,
Spea man’s ho).
Wi h he TF in he same posi ion, we hen deli e ed 405-nm ligh ac oss di e en posi ions
along he TF o he b ain slice while changing he lase inpu angle, ac ing on he con olling
ol age Vo a gal anome e mi o -based scanning sys em.20 Fo each inpu angle and esul ing
emission pa e n, we collec ed an image o he emission o ligh in issue using a sCMOS came a
on a wide- ield pa h o he wo-pho on mic oscope [Fig. 2(b)]. In his way, we measu ed he
ligh illumina ion ields Iðx; y; VÞp oduced by he TF a e e y inpu ol age Vo he gal o
[Fig. 2(e)], whe e ðx; yÞdeno es a posi ion in issue. Then, we used he Iðx; y; VÞ ields as
a pixel-by-pixel mul iplica i e mask o es ima e he po ion o he ibe collec ion ield
Cðx; yÞin ol ed in luo escence exci a ion o di e en inpu angles [Fig. 2( )]. This led us
o calcula e a pho ome y s ack o images Pðx; y; VÞ ep esen ing he pho ome y collec ion
ields acco ding o
EQ-TARGET; emp:in alink-;sec3.2;117;123Pðx; y; VÞ¼Cðx; yÞ×Iðx; y; VÞ;
whe e ×indica ed a pixel-by-pixel mul iplica ion.
The images in he illumina ion ields s ack Iðx; y; VÞmeasu ed on he sCMOS came a
we e also used o loca e he cen oid o he a ea o issue ec ui ed a e e y inpu angle using
Fig. 1 (Con inued) line shows he median smoo hed signal (window size: 4). Top igh : co onal
b ain slice showing he la ibe pene a ion ack su ounded by Me hoxy-X04-s ained amyloid
pa hology. The whi e dashed line shows he pene a ion ack. Scale: 1 mm. Bo om igh : co e-
la ion analysis compa ing pho ome y and his ology dep h p o iles ( , Spea man’s ho). The black
line shows a i ed linea eg ession. (d) Summa y co ela ion coe icien s ac oss h ee di e en
implan si es ( wo-sample - es ). 5xFAD, n¼13 eco dings om six mice. WT, n¼11 eco dings
om se en mice. Squa es, males; ci cles, emales. (e) Classi ica ion pe o mance based on in i o
pho ome y-based dep h p o iles.
By on e al.: Dep h- esol ed ibe pho ome y o amyloid plaque signals. . .
Neu opho onics 035014-9 Jul–Sep 2025 •Vol. 12(3)
Alzheime ’s disease pa hology using s a e-o - he-a ape ed ibe -based pho ome y in eely
beha ing mice.
Niall McAlinden was awa ded his PhD in su ace and in e ace op ics om T ini y College
Dublin in 2010. Since hen, he has se ed a he Ins i u e o Pho onics a he Uni e si y o
S a hclyde. He in es iga es op ical and elec onic sys ems o in e acing wi h he b ain, p i-
ma ily ocusing on he de elopmen o mic oLED de ices o op ogene ics and pho ome y.
Filippo Pisano is an associa e p o esso a he Uni e si y o Padua, Depa men o Physics and
As onomy “G. Galilei”and an a ilia ed esea che a he Pado a Neu oscience Cen e . He
ecei ed his PhD om he Ins i u e o Pho onics, Uni e si y o S a hclyde, in 2017. His esea ch
g oup ocuses on he de elopmen o label- ee op ical ools and me hods o cap u e mul iple
ypes o signals om he cen al ne ous sys em.
Ma co Pisanello is CTO a Op ogeniX s. .l., a company ha o e s ape ed ibe p obes and
equipmen o op ogene ics. He ecei ed his PhD in ma e ials and s uc u es enginee ing and
nano echnologies om Uni e si à del Salen o and Is i u o I aliano di Tecnologia in 2017.
He cu en ly de elops he p oduc lineup o Op ogeniX, wi h pa icula emphasis on he op o-
elec onic ha dwa e and so wa e componen s.
Jacques Fe ei a is a PhD s uden a he Uni e si y o Eas Anglia, Uni ed Kingdom.
P e iously, he was a echnician in he Saka a lab a he Uni e si y o S a hclyde, Uni ed
Kingdom, whe e he p o ided echnical suppo and insigh o lab membe s. His cu en esea ch
in es iga es he ole o a no el C. elegans p o ein in he un olded p o ein esponse pa hway o
he endoplasmic e iculum.
Cinzia Mon ina o g adua ed in biological sciences and ob ained a p o essional biology quali-
ica ion. In Oc obe 2023, she ecei ed he PhD in biological and en i onmen al sciences and
echnologies om he Uni e si y o Salen o. She is cu en ly a pos doc o al esea che a Is i u o
I aliano di Tecnologia, Cen e o Biomolecula Nano echnologies (IIT@CBN). She ocuses on
mul i unc ional neu al in e aces a ge ing deep b ain egions wi h esea ch in e es s in ibe
pho ome y, wo-pho on mic oscopy, and Raman spec oscopy.
Kei h Ma hieson is a p o esso a he Uni e si y o S a hclyde, Uni ed Kingdom. He holds
a 10-yea awa d om he Royal Academy o Enginee ing as pa o hei Chai in Eme ging
Technologies scheme, wi h a ocus on neu o echnology. His esea ch is ocused on he de elop-
men o de ices o in e ace wi h neu al sys ems. These a e mic o ab ica ed, op oelec onic,
implan able de ices o eco d and ac i a e neu al ac i i y wi hin he b ain.
Massimo De Vi o io is a ull p o esso a he Technical Uni e si y o Denma k (DTU) and
p incipal in es iga o o he Cen e o Biomolecula Nano echnologies o he Is i u o I aliano
di Tecnologia in Lecce, I aly, whe e he has se ed as a di ec o o he pas 10 yea s. His esea ch
ac i i y deals wi h he de elopmen o science and echnology applied o nanopho onics, nano-
elec onics, and nano and mic o elec omechanical sys ems (NEMS/MEMS) o senso s and
ac ua o s o he human body in heal hca e applica ions.
Fe uccio Pisanello is he coo dina o o he Cen e o Biomolecula Nano echnologies a he
I alian Ins i u e o Technology, whe e he leads he esea ch uni on mul i unc ional neu al
in e aces o deep b ain egions. His s udy ocuses on de eloping ad anced neu o echnologies
o mul imodal sensing and ac ua ion o neu al ac i i y. He has coo dina ed na ional and
in e na ional esea ch ini ia i es and au ho ed o e 70 pee - e iewed publica ions in pee - e iewed
jou nals.
Shuzo Saka a is a p o esso a he Uni e si y o S a hclyde, Uni ed Kingdom. He ea ned his
PhD in biophysics om Kyo o Uni e si y in 2002. His esea ch g oup applies a wide ange o
in i o app oaches o in es iga e s a e-dependen and cell- ype-speci ic in o ma ion p ocessing
in he b ain. He is also a scien i ic ad iso y boa d membe o Neu oVLC. The opic om his
pe spec i e is no ela ed o his company.
By on e al.: Dep h- esol ed ibe pho ome y o amyloid plaque signals. . .
Neu opho onics 035014-16 Jul–Sep 2025 •Vol. 12(3)
