Integrity and quantity of salivary cell-free DNA as a potential molecular biomarker in oral cancer: A preliminary study

Vol.:(0123456789)
1 3
Mic obial Ecology (2023) 86:2655–2665
h ps://doi.o g/10.1007/s00248-023-02266-y
RESEARCH
F equen Pa asi ism o Apis melli e a byT ypanosoma ids
inGeog aphically Isola ed A eas wi hRes ic ed Beekeeping
Mo emen s
DanielAguado‑López1· Ca olinaBa olomé2· AnaRi aLopes3,4· Do aHen iques3,4· Sa aKa a iSegu a5·
XulioMaside2· M.AlicePin o3,4· Ma ianoHiges1· RaquelMa ín‑He nández1,6
Recei ed: 10 Ap il 2023 / Accep ed: 7 July 2023 / Published online: 22 July 2023
© The Au ho (s), unde exclusi e licence o Sp inge Science+Business Media, LLC, pa o Sp inge Na u e 2023
Abs ac
T ypanosoma ids o m a g oup o high p e alence p o ozoa ha pa asi ise honey bees, wi h Lo ma ia passim as he p edominan
species wo ldwide. Howe e , he knowledge abou he ecology o ypanosoma ids in isola ed a eas is limi ed. The Po uguese
a chipelagos o Madei a and Azo es p o ide an in e es ing se ing o in es iga e hese pa asi es because o hei geog aphic
isola ion, and because hey ha bou honey bee popula ions de oid o wo majo enemies: Va oa des uc o and Nosema ce anae.
Hence, a o al o 661 honey bee colonies om Madei a and he Azo es we e analysed using di e en molecula echniques,
h ough which we ound a high p e alence o ypanosoma ids despi e he isola ion o hese islands. L. passim was he p e-
dominan species and, in mos colonies, was he only one ound, e en on islands ee o V. des uc o and/o N. ce anae wi h
se e e es ic ions on colony mo emen s o p e en he sp ead o hem. Howe e , islands wi h V. des uc o had a signi ican ly
highe p e alence o L. passim and, con e sely, islands wi h N. ce anae did no shown any signi ican co ela ion wi h he
ypanosoma id. C i hidia bombi was de ec ed in Madei a and on h ee islands o he Azo es, almos always coinciden wi h L.
passim. By con as , C i hidia melli icae was no de ec ed in any sample. A high- h oughpu sequencing analysis dis inguished
wo main haplo ypes o L. passim, which accoun ed o 98% o he o al sequence eads. This wo k sugges s ha L. passim and
C. bombi a e pa asi es ha ha e been associa ed wi h honey bees p eda ing he sp ead o V. des uc o and N. ce anae.
Keywo ds Honey bee· T ypanosoma ids· Lo ma ia passim· PCR· Sequencing· Va oa des uc o
In oduc ion
T ypanosoma idae a e a g oup o pa asi ic p o ozoa ha
can in ec a wide ange o o ganisms, om plan s o insec s
and e eb a es [1, 2]. In ecen yea s, pa icula in e es has
ocused on hei in ec ion o honey bees because o hei
possible in ol emen in inc eased colony mo ali y. The i s
ypanosoma id desc ibed in honey bees was C i hidia mel-
li icae [3] and, subsequen ly, C i hidia bombi was iden i ied
in bumble bees [4]. These pa asi es we e no gi en special
a en ion un il he beginning o he wen y- i s cen u y
[5–11] and un il hen, he ypanosoma ids ound in honey
bees we e hough o be limi ed o C. melli icae in honey
bees and o C. bombi in bumble bees. Howe e , he ad ances
in molecula echniques led o he disco e y o a new spe-
cies: Lo ma ia passim [8], whose exis ence had p e iously
been sugges ed om gene ic da a [12]. The e-e alua ion o
all he molecula da a ob ained in p e ious wo ks showed
ha many sequences we e inco ec ly assigned o C. melli-
icae and, in ac , i became clea ha L. passim was he p e-
dominan species wo ldwide [13–15]. Thus, L. passim seems
o ha e been p esen in honey bee colonies long be o e i s
desc ip ion, al hough he e is no scien i ic e idence as o
whe he i is a no el honey bee pa asi e ha ecen ly sp ead
wo ldwide, simila o Va oa des uc o and Nosema ce a-
nae, o whe he i is an old pa asi e.
Due o he geog aphic isola ion o he hos popula ions,
islands o e an ideal s age o assess hese hypo heses and
s udy he ecology and dynamics o in asi e pa hogen colo-
nisa ion. Ye , he e a e e y ew ypanosoma id s udies on
islands, and hese ha e been limi ed o he Paci ic, includ-
ing Japan [10, 16], New Zealand [15, 17, 18] and Hawaii
[19]. The e o e, su eys om islands a o he geog aphical
* Raquel Ma ín-He nández
[email p o ec ed]
Ex ended au ho in o ma ion a ailable on he las page o he a icle
2656 D.Aguado-López e al.
1 3
la i udes would shed addi ional ligh as o whe he hese
mic oo ganisms a e ecen o ancien honey bee pa asi es.
To his ega d, he Po uguese a chipelagos o he Azo es
and Madei a a e loca ed in he A lan ic Ocean, 1400km and
900km om he Eu opean mainland, Lisbon, espec i ely.
The i s in oduc ion o honey bees on hese islands da es
back o he six een h cen u y. These we e aken by he Po -
uguese se le s, o by la e inhabi an s, when he islands
we e used as s opo e s on he way o he Ame icas [20–22].
Acco ding o he la es epo o he Po uguese Bee Heal h
P og amme [23], he e a e cu en ly mo e han 10,000 honey
bee colonies in Madei a and a ound 8000 in he Azo es. The
se e e es ic ions on he impo a ion o honey bees imposed
since 2007 make he Azo es a chipelago epidemiologically
unique, wi h six o he nine islands emaining ee o V.
des uc o . Rema kably, in addi ion o being one o he ew
places in he wo ld ee o he mi e [24], wo o he Azo ean
islands (Flo es and San a Ma ia) ha e also escaped he
wo ldwide sp ead o he mic ospo idium N. ce anae [25].
The e o e, he Azo ean a chipelago p o ides an incompa-
able se ing o s udying pa asi e ecology, as i is home o
all kinds o pa asi e combina ions. In ligh o his, we con-
duc ed a su ey o ypanosoma ids in he Azo es as well
as on Madei a, which was ea lie colonised by V. des uc-
o . The ul ima e goal was o es ablish he p e ailing ypa-
nosoma id species and also o asce ain whe he he queen
ma ke ing and colony mo emen ha ha e been implica ed
in he sp ead o V. des uc o ha e also in luenced he dis-
ibu ion o ypanosoma ids ac oss he islands. To ha end,
we employed molecula me hods on a c oss-sec ional honey
bee sample ha allow de ec ion o an a ay o ypanosoma-
ids, including C. melli icae and L. passim, which ypically
pa asi ize honey bees, and also species such as C. bombi,
C i hidia expoeki and C i hidia acan hocephali, mo e a ely
associa ed wi h honey bees [26].
Me hods
Su ey andSample Collec ion
This c oss-sec ional s udy is pa o a la ge su ey con-
duc ed be ween 2014 and 2015 o desc ibe he p e alence
o pa hogens in he Azo es and Madei a a chipelagos. The
su ey was ca ied ou ollowing he design desc ibed else-
whe e [25, 27] and in acco dance wi h he numbe o colo-
nies on each island egis e ed in 2013, wi h an expec ed
pa hogen p e alence o 15%, a p ecision a e o 10% and a
con idence le el o 95%. Abou 150 adul honey bee wo k-
e s we e collec ed om 159 geo e e enced apia ies o he
Azo es in he Summe o 2014 and 2015, and om 23 geo-
e e enced apia ies o Madei a in he Sp ing o 2014. Sam-
ples we e collec ed om h ee andom colonies, and in a
ew cases om wo o ou colonies pe apia y, esul ing in
a o al o 483 samples o he Azo es and 89 o Madei a.
In addi ion, he islands o Faial, São Jo ge, San a Ma ia
and Te cei a we e e-sampled in he Summe o 2020 (89
colonies in 34 apia ies). In his la e sampling, each sample
comp ised 20–30 wo ke s. Samples om he Azo es we e
shipped ali e o he Cen o de In es igação de Mon anha
(CIMO, Po ugal) and hen sen on d y ice o he Cen o
de In es igación Apícola y Ag oambien al (CIAPA, Spain).
Samples om Madei a we e collec ed in e hanol and s o ed
a − 20°C un il shipping o CIAPA. All samples we e kep
a − 80°C o u he analysis.
DNA Ex ac ion
Fo each colony sampled in he Azo es and Madei a in
2014–2015, a subsample o 120 wo ke s was selec ed and
p ocessed as desc ibed p e iously [25]. In b ie , he honey
bees we e mace a ed in 50% AL bu e (Qiagen®) and cen-
i uga ed a 3000 pm o 10min o ob ain supe na an
and sedimen ha we e p ocessed sepa a ely. The sedimen
was esuspended in 3mL o milli-Q wa e and he DNA
was ex ac ed as desc ibed p e iously [25]. Only o he
2014–2015 Azo ean samples, 400 µL o each supe na an
was ans e ed o a 96-well pla e (Deepwell, Eppendo ),
ea ed wi h 15 µL o p o ease (Qiagen®) and incuba ed a
70°C o 10min. A e wa ds, he DNA was ex ac ed as
indica ed be o e [25].
The samples collec ed in 2020 in he Azo es we e p o-
cessed simila ly, bu in his case he 20–30 honey bees we e
mace a ed in 5mL o au ocla ed milli-Q wa e in a S om-
ache 80. In hese samples, no phase sepa a ion was made
so ha 400 µL o each mace a e was di ec ly ans e ed o
a 96-well Collec ion Mic o ube pla e (Qiagen®) wi h glass
beads (2mm diame e , Sigma) and shaken (3min, 30Hz) in
a TissueLyse (Qiagen®). Then, 150 µL o each sample was
dispensed in o a pla e wi h 150 µL o 50% AL bu e and 15
µL o p o ease. A e incuba ion, he DNA was ex ac ed as
abo e men ioned [25].
Ex ac ed nega i e con ols we e included in all he p o-
cessing and DNA ex ac ion s eps (one e e y 20 samples)
and p ocessed in pa allel. All samples we e s o ed a − 80°C
un il u he analysis.
Sedimen Ve sus Supe na an
Samples om he Azo es om 2014 o 2015 we e i s
p ocessed o he de ec ion o o he pa hogens in a di e -
en s udy [25] and, due o hei alue, we e eused in his
s udy. Since sedimen and supe na an we e a ailable and
i was no known which phase was he bes o de ec ing
hese pa asi es a e honey bee mace a ion, DNA ob ained
om he wo phases was analysed sepa a ely o de e mine
2657
F equen Pa asi ism o Apis melli e a byT ypanosoma ids inGeog aphically Isola ed A eas…
1 3
which was he bes subs a e o he de ec ion o T ypa-
nosoma idae. Fo his pu pose, DNA was ex ac ed om
he sedimen and supe na an o each sample, as abo e,
and analysed in pa allel by PCR, in o de o compa e he
numbe o posi i e samples ob ained in each phase.
T ypanosoma idae De ec ion
Polyme ase chain eac ion (PCR) was used o de ec he
T ypanosoma idae in all he samples (sedimen om
Madei a samples, supe na an and sedimen om he
2014–2015 Azo ean samples, and mace a e om he 2020
Azo ean samples) wi h he T yp RPB1 p ime s desc ibed
elsewhe e [26] and shown in Table1. These p ime s a ge
he DNA-dependen RNA polyme ase I ( pb1) gene and
hey ampli y a agmen o 283bp om all T ypanosoma-
idae species ha ha e been de ec ed in honey bees o da e.
PCRs we e pe o med in a 25-µL olume con aining
13.25 µL o H2O, 5 µL o 5X Phusion™ HF bu e , 0.5 µL
o a 10mM dNTP mix, 2.5 µL o each p ime (5µM), 0.25
µL o Phusion™ DNA Polyme ase (The mo Fishe ) and
1 µL o he DNA empla e. The PCR empe a u e p o ile
was se acco ding o he manu ac u e ’s ins uc ions, and
consis ed o an ini ial dena u a ion a 98°C o 30s, ol-
lowed by 45 cycles a 98°C o 10s, 62.2°C o 30s and
72°C o 10s, and a inal ex ension o 8min a 72°C.
PCRs we e ca ied ou in a Mas e cycle ® ep g adien S
(Eppendo ) and he esul ing amplicons we e analysed
in a QIAxcel Ad anced Sys em (Qiagen®), s o ing he
emaining PCR p oduc a − 20°C o u he p ocessing.
Ex ac ion and PCR nega i e con ols and a posi i e con-
ol we e included in all he analyses and un in pa allel. In
he case o 2014–2015 Azo ean samples, as bo h sedimen
and supe na an we e analysed, a sample was conside ed
posi i e i he e was ampli ica ion in a leas one o he
phases.
High‑Th oughpu Sequencing
A subse o 91 posi i e samples ha we e posi i e o T ypa-
nosoma idae we e sequenced on he Illumina MiSeq pla -
o m oge he wi h nega i e con ols. The PCR p oduc o
he posi i e samples was quan i ied in a NanoD op™ 2000
Spec opho ome e (The moScien i ic™) o de e mine he
DNA concen a ion and sen ozen o Uni e sidade de San-
iago de Compos ela (CIMUS, Spain) o lib a y p epa a ion
and sequencing.
Amplicons om each sample we e used as inpu o p e-
pa e he lib a y using he KAPA Hype P ep ki (Roche
Sequencing Solu ions Inc.), ollowing he manu ac u e ’s
p o ocol di ec ly om he end- epai and A- ailing s ep.
Lib a y pools we e no malised o a concen a ion o 4nM
and loaded a a concen a ion o 12pM on an Illumina
MiSeq ins umen o 1 × 300bp single-end sequencing
(Flowcell Nano V2, 2 × 150bp).
The eads gene a ed in he sequencing un we e de-mul-
iplexed acco ding o he ba codes assigned o each sample,
and hen p ocessed wi h as p [29] o adap o emo al and
quali y il e ing. The eads we e hen o ganised in o indi-
idual iles ha con ained he numbe and sequence hap-
lo ypes de ec ed o each amplicon in a sample, as well as
he numbe o eads in each di ec ion ( o wa d and e e se).
Sequences wi h less han 5 eads we e disca ded. Subse-
quen ly, he iles we e con e ed o as a o ma o sequence
alignmen wi h MACSE 2.05 (Mul iple Alignmen o Cod-
ing SEquences), a p og am ha aligns p o ein-coding gene
da ase s wi hou dis up ing he unde lying codon s uc u e
[29]. The e e ence sequences o he pb1 we e ob ained
om GenBank o L. passim (MT558272.1 and LT976801.
1), C. melli icae (MT558227.1 and MT558204.1), C.
bombi (MT558162.1 and MT558134.1), C. acan hocephali
(MW28878781.1) and a new T ypanosoma idae species
(T ypanosoma idae sp. MN038411.1), ecen ly desc ibed
[26]. Sequences we e isualised wi h BioEdi [30] and hose
Table 1 P ime s and p obes
used o he molecula analysis
o he honey bee samples
Re e ences Ta ge P ime s Sequence 5ʹ–3ʹ
[26] T ypanosoma idae T yp RPB1-F1 GTG GCT GGA YCT GTG GGA GC
T yp RPB1-R1 GCC RTT GAT GAA CTT CGC CAC
[28]C. melli icae qCmell_Cy b_F TTT TGC CAT GCA CTA TGA TGTCT
qCmell_Cy b_R AAC CTA TTA CAG GCA CAG TTG CTA AA
qCmell_Cy b_P 6FAM-ATT GAG GAT TAA CAG TGT TTAGT-BHQ1
L. passim qLpass_TOPII_F GGC CAT GGA AAT ACT CGA GTCT
qLpass_TOPII_R ACC TTG CCT TCC TTC TTG AGATT
qLpass_TOPII_P 6FAM-CCT CGA CACGC + T + TA + GT-BHQ1
C. bombi qCbom_RPB1_F TGG TGG GTG CGA TTA CGA A
qCbom_RPB1_R TCA TTG AAG ATG ACG TGG ATA AGC
qCbom_RPB1_P 6FAM-CGT TGT CGG CGC CG-BHQ1
2658 D.Aguado-López e al.
1 3
con aining indels o s op codons we e emo ed om he
inal da ase .
Species‑Speci ic De ec ion
All T ypanosoma idae-posi i e samples ha we e no
sequenced by high- h oughpu sequencing (HTS) we e ana-
lysed by eal- ime qPCR (qPCR) o he species-speci ic
de ec ion o L. passim, C. melli icae and C. bombi using he
p ime s desc ibed elsewhe e [28] and shown in Table1. The
qPCR was pe o med on DNA ob ained om Madei a sam-
ples, om bo h sedimen and supe na an o he 2014–2015
Azo ean samples, and om samples wi hou phase sepa a-
ion om he 2020 Azo ean samples. Nega i e and posi i e
con ols we e also es ed in pa allel in all PCRs. As o T yp-
anosoma idae de ec ion in he 2014–2015 Azo ean samples,
a sample was conside ed posi i e i he e was ampli ica ion
om ei he o he wo phases. Amplicons om samples
ha ga e a nega i e esul by he speci ic qPCR bu a posi-
i e esul by he s anda d PCR ampli ica ion wi h he T yp
RPB1 p ime s we e pu i ied wi h QIAquick (Qiagen®) and
Sange sequenced in bo h di ec ions (Genomic Uni , Uni-
e sidad Complu ense, Spain) on a 3730 Gene ic Analyze
(Applied Biosys ems). The sequences we e checked manu-
ally wi h BioEdi [30] and compa ed wi h he sequences
downloaded om GenBank using BLAST.
The geog aphical dis ibu ion o he colonies was com-
piled and plo ed o each island using he A cGIS desk op
so wa e [31].
S a is ical Analysis
Chi-squa e es s we e pe o med o compa e he sensi i i y
o T ypanosoma idae de ec ion in he sedimen and supe na-
an phases. Gene alised linea models wi h a binomial am-
ily and a logi link unc ion we e used o de e mine whe he
he p esence o V. des uc o o N. ce anae in luenced he
p e alence o L. passim a he island scale, including island
as co a iable. In he la e case, only da a om samples om
he Azo es collec ed in 2014–2015 we e used, as his sam-
pling was conduc ed and analysed homogeneously.
Resul s
Compa ison Be ween Sedimen andSupe na an
To de e mine he bes phase o T ypanosoma idae de ec ion,
sedimen and supe na an analysis was pe o med on 477
Azo ean samples collec ed in 2014–2015 (Table2; six sam-
ples ou o 483 had insu icien supe na an o his assay).
The supe na an p oduced he highes numbe o posi i e
samples (55.34%) as compa ed o he sedimen (50.73%),
al hough no s a is ically signi ican di e ences we e ound
be ween bo h phases (X2=2.04, gl=1; p=0.15). No ypano-
soma id DNA was de ec ed in any nega i e con ols.
P e alence o T ypanosoma idae
The p e alence o T ypanosoma idae is shown in Table3.
As he Madei a, Azo es 2014–2015 and Azo es 2020 sam-
ples we e p ocessed di e en ly, no compa isons we e made
be ween hose 3 g oups. In Madei a, he p e alence was
66.3%, whe eas in he Azo es was 72% in 2014–2015, wi h
he highes numbe o in ec ed colonies de ec ed on Pico
(92%) and he lowes on G aciosa (22.7%). In he la e sam-
pling pe iod (2020), 31.5% o he colonies we e posi i e,
Table 2 Numbe (and pe cen age) o T ypanosoma idae-posi i e ( +)
and T ypanosoma idae-nega i e ( −) samples de ec ed on he supe -
na an and sedimen phases ob ained om he honey bee mace a es
Supe na an ( +) Supe na an ( −) To al
Sedimen ( +) 161 (33.75%) 81 (16.98%) 242 (50.73%)
Sedimen ( −) 103 (21.59%) 132 (27.67%) 235 (49.27%)
To al 264 (55.34%) 213 (44.65%) 477 (100%)
Table 3 Numbe o colonies and apia ies sampled in each island,
numbe o colonies posi i e o T ypanosoma idae and co esponding
p e alence (%). 1Va oa- ee islands; 2Nosema- ee islands
Apia ies
(N)
Colonies
(N)
Posi i e
colonies
(N)
P e alence
(%)
2014–2015
Madei a 23 89 59 66.3
Azo es
San a Ma ia1, 2 19 57 48 84.2
São Miguel130 105 84 80.0
São Jo ge117 37 19 51.4
Faial 20 60 43 71.7
Pico 25 75 69 92.0
G aciosa17 22 5 22.7
Te cei a126 80 43 53.8
Flo es215 47 37 78.7
To al 159 483 348 72.0
2020
San a Ma ia1, 2 12 28 12 42.9
São Jo ge110 30 5 16.7
Faial 2 8 2 25.0
Te cei a110 23 9 39.1
To al 34 89 28 31.5
To al 216 661 435
2659
F equen Pa asi ism o Apis melli e a byT ypanosoma ids inGeog aphically Isola ed A eas…
1 3
wi h he la ges pe cen age de ec ed on San a Ma ia (42.9%)
and he lowes on São Jo ge (16.7%).
Species Iden i ica ion
High‑Th oughpu Sequencing
A o al o 91 samples iden i ied as posi i e o T ypanoso-
ma idae by he PCR assay we e sequenced on he Illumina
MiSeq pla o m. The sequencing un p oduced 1,385,021
aw eads, which yielded 1,208,665 T ypanosoma idae
sequences a e il e ing, ep esen ing an a e age o 13,282
sequences pe sample. S ikingly, iden i ica ion o all
sequence eads e ealed ha L. passim was he only species
p esen in he da ase . Two main haplo ypes we e ound in
all samples and on all islands and accoun ed o abou 98%
o he sequences ound (Table4) and hese ma ched he e -
e ence sequences MT558272.1 and LT976801.1 (he ea e
e e ed o as Lp1 and Lp2, espec i ely) downloaded om
GenBank. These wo haplo ypes we e ound dis ibu ed
wi hin each island wi hou any de ined geog aphical pa e n
and hey di e ed by a single nucleo ide in he amplicon a
posi ion 29, wi h Lp1 ha ing a T and Lp2, a C (Supp. Fig-
u e1). On a e age, 72.9% o he eads om all he islands
co esponded o he Lp2 haplo ype, whe eas 25.3% o he
sequences co esponded o he Lp1 haplo ype. This di -
e ence was mos no able in samples collec ed in Madei a
and in samples collec ed in 2014–2015 om all islands
o cen al g oup, whe e mo e han he 70% o sequences
co esponded o he Lp2 haplo ype (Table4; Fig.1). The
emaining 2% o sequences we e ound a a e y low e-
quency and appea ed o be a ian s om he majo i y hap-
lo ypes wi h addi ional single nucleo ide changes and did no
ma ch any sequence a ailable in GenBank. The sequences
o he haplo ypes ound (bo h Lp1 and Lp2 and hose a low
equencies) we e deposi ed in GenBank (accession numbe s
OR117383-OR117469).
Real‑ ime qPCR Species Con i ma ion
The es o he T yp-RPB1 posi i e samples (n=344) we e
analysed by qPCR o de ec he p esence o L. passim,
C. melli icae and/o C. bombi. A leas one o hese spe-
cies was iden i ied in 341 samples, whe eas h ee samples
( om Madei a, San a Ma ia and São Miguel) we e nega i e
(Table5). No ampli ica ion was obse ed in any o he nega-
i e con ols.
The qPCR analysis con i med ha L. passim was he
p edominan bu no he only species on Madei a and in
he Azo es, as C. bombi was also ound on some islands
(Table5). In he i s sampling pe iod, in ec ion by L. pas-
sim was iden i ied alone in 64.2% o he posi i e samples on
Madei a, while C. bombi was ound in 34% o he posi i e
samples, always oge he wi h L. passim (Table5). On São
Jo ge, Pico, G aciosa, San a Ma ia and Te cei a, L. passim
was he only species de ec ed, whe eas C. bombi was ound
as a mono-in ec ion in one colony on Flo es (3.7%). Ye , on
São Miguel and Faial, his species was always associa ed
wi h L. passim in ec ion. In he samples collec ed in 2020, L.
passim was he only T ypanosoma idae de ec ed (Table5).
Finally, C. melli icae was no ound in any o he samples
analysed h oughou he s udy.
Sange sequencing o he T yp RPB1 p ime amplicons
o he h ee nega i e samples a e qPCR showed 100%
homology wi h he MT558162.1 sequence o C. bombi in
he Madei a sample, in he San a Ma ia sample he sequence
was iden ical o Lp2, while in he São Miguel sample, wo
peaks o C/T a posi ion 29 o he amplicon, co esponding
o he Lp1 and Lp2 haplo ypes, we e easily dis inguishable.
To al Occu ence o T ypanosoma idae Species onMadei a
and heAzo es
The occu ence o each T ypanosoma idae species calcu-
la ed by compiling all he esul s ob ained using he di e en
echniques (HTS, qPCR and Sange sequencing) is shown
in Table6 and Fig.1. On Madei a, he p opo ion o colo-
nies in which T ypanosoma idae we e de ec ed was 66.3%,
wi h L. passim being he only species de ec ed in 44.9%, C.
bombi in 1.1% and bo h species concomi an ly in 20.2% o
he samples analysed. In he 2014–2015 samples om he
Azo es, 348 colonies we e posi i e o T ypanosoma idae
Table 4 A e age numbe o o al and alid eads (sequences wi h < 5
eads we e emo ed), gene a ed by he MiSeq pla o m o 91 T ypa-
nosoma idae-posi i e samples, and p opo ion o he wo main L. pas-
sim haplo ypes (Lp1 and Lp2) de ec ed on each island. 1Va oa- ee
islands; 2Nosema- ee islands
NTo al eads Valid eads % Lp1 % Lp2
2014–2015
Madei a 6 12,678 11,509 34.2 64.1
Azo es
San a Ma ia1, 2 10 15,809 14,547 39.2 59.0
São Miguel110 12,940 11,812 39.5 58.8
São Jo ge17 12,982 11,700 6.5 91.0
Faial 10 12,886 11,883 14.5 83.6
Pico 10 12,642 11,652 14.9 83.2
G aciosa14 12,827 11,725 27.9 70.9
Te cei a19 12,204 11,152 9.0 89.3
Flo es210 13,256 12,137 47.6 50.8
2020
San a Ma ia1, 2 5 14,321 13,146 37.7 60.5
São Jo ge15 13,459 12,490 19.8 78.4
Te cei a15 13,261 12,116 12.8 84.7
To al 91 13,282 12,168 25.3 72.9

2660 D.Aguado-López e al.
1 3
(72.0%). L. passim was ound on all he islands, and i was
he only species de ec ed in 71.2% o samples. C. bombi was
a ely ound in he Azo es, being he only species de ec ed
in one colony on Flo es (0.2%) and occu ed concu en ly
wi h L. passim on São Miguel and Faial (0.6%). In he 2020
samples, all he colonies ha we e posi i e o T ypanoso-
ma idae (31.5%) we e also posi i e o L. passim, and his
was he only species ound in his sampling pe iod (Table6).
Finally, as p e iously men ioned, C. melli icae was no
de ec ed in any o he samples es ed.
As L. passim was he main species ound, he in luence
o V. des uc o and N. ce anae was only s udied o his
ypanosoma id. Thus, he p e alence o L. passim was sig-
ni ican ly highe on he Azo es islands (2014–2015 samples)
wi h V. des uc o (Table7, eg ession coe icien = 0.77; z
alue = 3.41; p < 0.01) han on hose wi hou he mi e. In
Fig. 1 Geog aphical dis ibu-
ion o C. bombi and L. passim
(and i s haplo ypes) in he
Azo es a chipelago (a) and
Madei a (b). indica es he
islands whe e V. des uc o is
p esen . indica es he islands
whe e N. ce anae is p esen .
The ed colou in sec o cha s
indica es he pe cen age o
sequences ha ma ched wi h
Lp1 haplo ype while he pink
colou indica es he pe cen age
o sequences ha ma ched wi h
Lp2 haplo ype. C. bombi + L.
passim;C. bombi;L. pas-
sim;Nega i e sample
2661
F equen Pa asi ism o Apis melli e a byT ypanosoma ids inGeog aphically Isola ed A eas…
1 3
con as , on islands whe e N. ce anae was es ablished, he
p e alence o L. passim was lowe (Table7; eg ession coe -
icien = − 0.56; z alue = − 2.03; p < 0.01). Howe e , when
island was included as co a iable, he e ec o N. ce anae on
he p esence o L. passim became s a is ically no signi ican
( eg ession coe icien = − 0.29; z alue = − 0.66; p = 0.51)
while he signi icance was kep o V. des uc o ( eg ession
coe icien = 1.06; z alue = 2.15; p = 0.03).
Discussion
The aim o his s udy was o assess he s a us o T ypano-
soma idae species in honey bee colonies in Madei a and he
Azo es, and o de e mine whe he he p esence o V. des uc o
and N. ce anae on some islands in luenced hei dis ibu ion.
Ou esul s show ha islands wi h he mi e had a signi ican ly
Table 5 Pe cen age o each
T ypanosoma idae species
de ec ed in posi i e samples.
ND, no de ec ed. 1Va oa- ee
islands; 2Nosema- ee islands
NOnly L. passim Only C. mel-
li icae Only C. bombi Co-occu ence L.
passim/C. bombi
2014–2015
Madei a 53 64.2 ND ND 34
Azo es
San a Ma ia1, 2 38 97.4 ND ND ND
São Miguel174 95.9 ND ND 2.7
São Jo ge112 100 ND ND ND
Faial 33 97 ND ND 3
Pico 59 100 ND ND ND
G aciosa11 100 ND ND ND
Te cei a134 100 ND ND ND
Flo es227 96.3 ND 3.7 ND
To al 278 97.8 ND 0.4 1.1
2020
San a Ma ia1,2 7 100 ND ND ND
Faial 2 100 ND ND ND
Te cei a14 100 ND ND ND
To al 13 100 ND ND ND
Table 6 P e alence o
T ypanosoma idae (%) de ec ed
ei he alone o simul aneously
on each island g ouping he
da a om all he ools used
(con en ional PCR, HTS,
qPCR and Sange sequencing).
Numbe o samples analysed
as shown in Table3. ND, no
de ec ed. 1Va oa- ee islands;
2Nosema- ee islands
T ypanoso-
ma idae
Only L. passim Only C.
melli icae Only C. bombi Co-occu ence L.
passim/C. bombi
2014–2015
Madei a 66.3 44.9 ND 1.1 20.2
Azo es
San a Ma ia1, 2 84.2 84.2 ND ND ND
São Miguel180.0 78.1 ND ND 1.9
São Jo ge151.4 51.4 ND ND ND
Faial 71.7 70.0 ND ND 1.7
Pico 92.0 92.0 ND ND ND
G aciosa122.7 22.7 ND ND ND
Te cei a153.8 53.8 ND ND ND
Flo es278.7 76.6 ND 2.1 ND
To al 72.0 71.2 ND 0.2 0.6
2020
San a Ma ia1, 2 42.9 42.9 ND ND ND
São Jo ge116.7 16.7 ND ND ND
Faial 25.0 25.0 ND ND ND
Te cei a139.1 39.1 ND ND ND
To al 31.5 31.5 ND ND ND
2662 D.Aguado-López e al.
1 3
highe p e alence o L. passim. O he s udies had epo ed an
associa ion be ween hem [32] e en de ec ing DNA o his
ypanosoma id in he aca i, which is s ill o unce ain biologi-
cal signi icance [33]. In con as , when he island was included
as co a iable, no e ec o N. ce anae on L. passim p e alence
was obse ed as i was p e iously epo ed by o he au ho s
[14, 34]. Howe e , a posi i e co ela ion be ween in ec ion le -
els in colonies wi h bo h pa asi es has been also desc ibed [35].
Despi e hese di e ences ela ed o he p esence o V.
des uc o and N. ce anae, i is clea ha ypanosoma ids
a e well es ablished in he Azo es and Madei a, so ha he
s ong es ic ions on he in oduc ion o honey bees on he
Va oa- ee islands o he Azo es ha e no p e en ed he
p esence o ypanosoma ids. Fu he mo e, L. passim is he
main species ound and he only iden i ied in mos o he
colonies analysed. In e es ingly, C. melli icae, he species
associa ed wi h honey bees un il he disco e y o L. passim
[8, 12], is no p esen in he e i o ies analysed o is below
ou de ec ion limi . Nei he C. acan hocephali no C. expoeki
we e de ec ed in his s udy, al hough hey could only ha e
been con i med by HTS as hey we e no analysed by qPCR.
This scena io whe e L. passim is he main ypanoso-
ma id species in honey bee colonies seems o be epea ed
wo ldwide [13, 14, 28, 32, 35–39]. Gi en he high p e a-
lence ound he ein and elsewhe e, i is likely ha L. passim,
a he han C. melli icae, was he p ima y species in ec ing
honey bees in s udies conduc ed p io o i s disco e y [9–12].
Indeed, he equencies obse ed and he absence o C. mel-
li icae in many a eas o he wo ld co obo a e he hypo hesis
ha L. passim is a common pa asi e o Apis melli e a and ha
i is cu en ly he dominan ypanosoma id species in adul
wo ke bees [8, 13, 28, 39, 40]. On he o he hand, C. bombi
was iden i ied in h ee islands and almos always coexis ed
wi h L. passim al hough i also appea ed alone in some colo-
nies. These esul s u he con i m p e ious indings [41, 42],
in which C. bombi was de ec ed in honey bees.
How ypanosoma ids en e ed he Azo es is s ill unclea
since he islands a e loca ed a om he mainland; he na u-
al a i al o bees is no possible, so a human-made in oduc-
ion was e y p obably. Al hough i could be ha Pico was
he poin o en y, as i hos s he highes numbe o ypa-
nosoma id-posi i e colonies and was also he poin o en y
o V. des uc o in o he Azo es, he da a om San a Ma ia,
whe e nei he V. des uc o no Nosema spp. a e p esen [25]
bu whe e L. passim was de ec ed in a la ge p opo ion o
he colonies, do no suppo his hypo hesis. The widesp ead
dis ibu ion o L. passim on he Azo ean islands ha emain
ee o V. des uc o s ongly sugges s ha his species has
been associa ed wi h A. melli e a since be o e he wo ldwide
sp ead o he mi e. A simila scena io was ound in Hawaii
a chipelago, in which s ill has some islands whe e he mi e
is no p esen and whe e L. passim has been de ec ed [19].
In New Zealand (Va oa- ee un il 2000), signi ican colony
losses began o be obse ed in 2014 and an analysis o he
su i ing colonies e ealed ha mos we e in ec ed wi h
L. passim [15, 17, 18]. On he o he hand, he p esence o
bumblebees in Madei a [43] and in he Azo es [44] could
explain he p esence o C. bombi in honey bees due o eed-
ing on common lo al esou ces [45–47].
Ou s udy suppo s ha bo h sedimen and supe na an
phases a e sui able o ypanosoma id de ec ion. Also, he use
o di e en echnologies was key o iden i y all he species p e-
sen in his a ea o Maca onesia. Thus, on he one hand, HTS
echnology allowed us o de ec he haplo ypes p esen bu no
all he species, as he cos o his ool limi ed he numbe o
samples analysed (n = 91). On he o he hand, he use o qPCR
se ed o complemen he analysis by allowing us o analyse
he emaining ypanosoma id-posi i e samples (n = 344),
which se ed o iden i y C. bombi p esen in lowe p e alence.
The e is li le da a a ailable on he popula ion gene ics o he
pb1 in he ypanosoma ids ha in ec honey bees o bumble
bees. Recen ly, his gene was analysed and gene ic polymo -
phisms we e obse ed be ween C. melli icae, C. bombi and
L. passim [26, 48, 49]. Ou esul s a e consis en wi h hose
ob ained o he RPB1 gene in hese h ee s udies, indica ing
ha he e we e wo co e haplo ypes di e ing in only one nucle-
o ide. I canno be uled ou ha he sequences ound a low
equencies co espond o ue a e haplo ypes, al hough u -
he expe imen al con i ma ion would be needed. O he genes
like 18S RNA, 28S RNA and i s-2 ha e been s udied o e alu-
a e in aspeci ic a ia ion, and hei sequences we e almos
iden ical be ween he di e en s ains o L. passim, wi h he
excep ion o he i ulence ac o gp63, which seems o ha e a
a iable s e ch be ween s ains [40]. The e o e, i is possible
ha he pb1 does no ha e su icien a ia ion o dis inguish
s ains, bu ha his may occu in o he genes.
In conclusion, despi e geog aphical isola ion, ypanoso-
ma ids equen ly in ec colonies in Madei a and he Azo es,
e en when he e a e es ic ions on honey bee mo emen s
o mi e- ee islands. L. passim is he p edominan T ypano-
soma idae species, occu ing as he only species in many
colonies and showing a highe p e alence on islands wi h V.
des uc o and lowe p e alence on islands wi h N. ce anae.
Addi ionally, C. bombi also occu s, o en co-occu ing wi h
L. passim. The de ec ion o L. passim and C. bombi in a eas
Table 7 Numbe (and pe cen age) o L. passim–posi i e ( +) and L.
passim–nega i e ( −) samples collec ed in 2014–2015 in he Azo es
dis ibu ed acco ding o whe he hey we e on islands whe e V.
des uc o o N. ce anae we e p esen ( +) o no ( −)
L. passim ( +) L. passim ( −) To al
V. des uc o ( +) 148 (81.32%) 34 (18.68%) 182 (100%)
V. des uc o ( −) 199 (66.11%) 102 (33.89%) 301 (100%)
N. ce anae ( +) 263 (69.39%) 116 (30.61%) 379 (100%)
N. ce anae ( −) 84 (80.77%) 20 (19.23%) 104 (100%)
2663
F equen Pa asi ism o Apis melli e a byT ypanosoma ids inGeog aphically Isola ed A eas…
1 3
whe e V. des uc o and N. ce anae a e no p esen sugges s
ha hey ha e been associa ed wi h A. melli e a o much
longe han o iginally hough .
Supplemen a y In o ma ion The online e sion con ains supplemen-
a y ma e ial a ailable a h ps:// doi. o g/ 10. 1007/ s00248- 023- 02266-y.
Acknowledgemen s The au ho s wish o hank V. Albendea, T. Co -
ales, M. Gaje o, C. Uce a and M. Buendía o he Honey Bee Pa hol-
ogy Labo a o y (CIAPA) and D. da Sil a (San iago de Compos ela
Uni e si y). We a e g a e ul o all e e ina ians and echnicians o
he “Di eção Regional da Ag icul u a e Desen ol imen o Ru al dos
Aço es” and “Di isão de Expe imen ação e Melho ia Ag ıcola, Di ecão
Regional de Ag icul u a da Madei a” who collec ed he samples ac oss
he Azo es and Madei a, namely F ank Aguia , Nuno Sal ado , Janyne
Sousa, I an cas o, Célia Mesqui a, Ana Jo ge, José Dias, Paulo Rico,
Ped o Leal, Vagne Paulos, Luis Xa ie , Luís Sil a, Ma ins Sil a, Ca -
los Gou eia, Ana Ca ina Coimb a, João Ramos, João A uda, Edga do
Melo, João Luís, Moniz da Pon e and Be a Co eia. Special hanks a e
due o Paula Viei a, F ank Aguia and Be a Co eia o coo dina ing
he sampling in he Azo es and Madei a.
Au ho Con ibu ion Concep ualiza ion: R.M.-H., M.A.P. and M.H.;
me hodology: all au ho s; o mal analysis: all au ho s; in es iga ion:
D.A-L., A.R.L. and S.K.S.; da a cu a ion: D.A.L.; w i ing—o iginal
d a p epa a ion: D.A-L. and R.M.-H.; w i ing— e iew and edi ing:
D.A-L., R.M.-H., C.B, X.M, M.A.P. and M.H.; supe ision: R.M.-H.;
p ojec adminis a ion and unding acquisi ion: R.M.-H and M.A.P; All
au ho s ha e ead and ag eed o he published e sion o he manusc ip .
Funding This wo k was unded by he Conseje ía de Educación,
Cul u a y Depo es, o he Jun a de Cas illa – La Mancha (Eu opean
Regional De elopmen Fund) P ojec No. SBPLY/19/180501/000334,
and h ough he p og am COMPETE 2020—POCI (P og ama
Ope acional pa a a Compe i idade e In e nacionalização) and FCT
(Fundação pa a a Ciência e a Tecnologia) in he amewo k o he
p ojec BeeHappy (POCI-01–0145-FEDER-029871). INCRECYT
p og am was unded by ESF/EC (Eu opean Social Funds). D.A-L
con ac was unded by he Minis e io de Asun os Económicos y
T ans o mación Digi al (G an No. PRE2018-084878, RTA2017-
00004-C02-01). A.R.L was unded by Fundação pa a a Ciência
e Tecnologia (FCT) h ough he indi idual esea ch g an SFRH/
BD/143627/2019. FCT p o ided inancial suppo by na ional unds
(FCT/MCTES) o CIMO (UIDB/00690/2020 and UIDP/00690/2020)
and SusTEC (LA/P/0007/2021).
Da a A ailabili y Lo ma ia passim sequences ob ained by HTS a e
a ailable a Genbank (NCBI) da a base. To al and alid eadsgene a ed
by he MiSeq pla o m a e a ailable a Supplemen a y Table1. O he
da ase s gene a ed du ing and/o analyzed du ing he cu en s udy
a e a ailable om he co esponding au ho upon easonable eques .
Decla a ions
Compe ing In e es s The au ho s decla e no compe ing in e es s.
Re e ences
1. Lopes HA (2010) T ypanosoma ids: odd o ganisms, de as a ing
diseases. Open Pa asi ol J 4:30–59. h ps:// doi. o g/ 10. 2174/ 18744
21401 00401 0030
2. Kau e A, Ellis J, S a k D, Ba a J (2017) The e olu ion o
ypanosoma id axonomy. Pa asi Vec o s 10:287. h ps:// doi.
o g/ 10. 1186/ s13071- 017- 2204-7
3. Lang idge DF, McGhee RB (1967) C i hidia melli icae n. sp.
an acidophilic ypanosoma id o he honey bee Apis melli e a.
J P o ozool 14:485–487. h ps:// doi. o g/ 10. 1111/j. 1550- 7408.
1967. b020 33.x
4. Lipa JJ, T iggiani O (1988) C i hidia bombi sp. n. a lagella ed
pa asi e o a bumble-bee Bombus e es is L. (Hymenop e a,
Apidae) 27:287–290
5. Meeus I, De G aa DC, Jans K, Smagghe G (2010) Mul iplex
PCR de ec ion o slowly-e ol ing ypanosoma ids and neog e-
ga ines in bumblebees using b oad- ange p ime s. J Appl
Mic obiol 109:107–115. h ps:// doi. o g/ 10. 1111/j. 1365- 2672.
2009. 04635.x
6. Schmid-Hempel R, Tognazzo M (2010) Molecula di e gence
de ines wo dis inc lineages o C i hidia bombi (T ypanosoma i-
dae), pa asi es o bumblebees. J Euka yo Mic obiol 57:337–345.
h ps:// doi. o g/ 10. 1111/j. 1550- 7408. 2010. 00480.x
7. Runckel C, Flenniken ML, Engel JC e al (2011) Tempo al analy-
sis o he honey bee mic obiome e eals ou no el i uses and
seasonal p e alence o known i uses, Nosema, and C i hidia.
PLoS One 6:e20656. h ps:// doi. o g/ 10. 1371/ jou n al. pone. 00206
56
8. Schwa z RS, Bauchan GR, Mu phy CA e al (2015) Cha ac e i-
za ion o wo species o T ypanosoma idae om he honey bee
Apis melli e a: C i hidia melli icae Lang idge and McGhee, and
Lo ma ia passim n. gen., n. sp. J Euka yo Mic obiol 62:567–583.
h ps:// doi. o g/ 10. 1111/ jeu. 12209
9. Ra oe J, Maha amo J, Meeus I e al (2013) Comp ehensi e bee
pa hogen sc eening in Belgium e eals C i hidia melli icae as a
new con ibu o y ac o o win e mo ali y. PLoS One 8:e72443.
h ps:// doi. o g/ 10. 1371/ jou n al. pone. 00724 43
10. Mo imo o T, Kojima Y, Yoshiyama M e al (2013) Molecula
de ec ion o p o ozoan pa asi es in ec ing Apis melli e a colonies
in Japan. En i on Mic obiol Rep 5:74–77. h ps:// doi. o g/ 10.
1111/j. 1758- 2229. 2012. 00385.x
11. Yang B, Peng G, Li T, Kadowaki T (2013) Molecula and phy-
logene ic cha ac e iza ion o honey bee i uses, Nosema mic o-
spo idia, p o ozoan pa asi es, and pa asi ic mi es in China. Ecol
E ol 3:298–311. h ps:// doi. o g/ 10. 1002/ ece3. 464
12. Cepe o A, Ra oe J, Gómez-Mo acho T e al (2014) Holis ic
sc eening o collapsing honey bee colonies in Spain: a case s udy.
BMC Res No es 7:649. h ps:// doi. o g/ 10. 1186/ 1756- 0500-7- 649
13. A ismendi N, B una A, Zapa a N, Va gas M (2016) PCR-speci ic
de ec ion o ecen ly desc ibed Lo ma ia passim (T ypanosoma-
idae) in Chilean apia ies. J In e eb Pa hol 134:1–5. h ps:// doi.
o g/ 10. 1016/j. jip. 2015. 12. 008
14. S e ano ic J, Schwa z RS, Vejno ic B e al (2016) Species-spe-
ci ic diagnos ics o Apis melli e a ypanosoma ids: a nine-yea
su ey (2007–2015) o ypanosoma ids and mic ospo idians in
Se bian honey bees. J In e eb Pa hol 139:6–11. h ps:// doi. o g/
10. 1016/j. jip. 2016. 07. 001
15. Hall RJ, P age H, Phi i BJ e al (2021) Apicul u al p ac ice and
disease p e alence in Apis melli e a, New Zealand: a longi udinal
s udy. J Apic Res 60:644–658. h ps:// doi. o g/ 10. 1080/ 00218 839.
2021. 19364 22
16. Yamamo o S, Nakamu a S, Nakayama K e al (2023) Molecula
de ec ion o Lo ma ia passim in honeybees in Japan. Pa asi ol In
93:102711. h ps:// doi. o g/ 10. 1016/j. pa in . 2022. 102711
17. McFadden AMJ, Mackay J, Bo owick O, Goodwin RM (2016)
In es iga ion o bee mo ali y in a beekeepe ope a ion in he
Co omandel dis ic . Su eill 43:9–11
18. Wa e s TL (2018) The dis ibu ion and popula ion dynamics o he
honey bee pa hogens C i hidia melli icae and Lo ma ia passim in
New Zealand. Disse a ion, Vic o ia Uni e si y o Welling on