This is a sel -a chi ed e sion o an o iginal a icle. This e sion
may di e om he o iginal in pagina ion and ypog aphic de ails.
Au ho (s):
Ti le:
Yea :
Ve sion:
Copy igh :
Righ s:
Righ s u l:
Please ci e he o iginal e sion:
CC BY 4.0
h ps://c ea i ecommons.o g/licenses/by/4.0/
The Riddle o How Fishe ies In luence Gene ic Di e si y
© 2023 by he au ho s. Licensee MDPI, Basel, Swi ze land
Published e sion
Sadle , Daniel E.; Wa s, Phillip C.; Uusi-Heikkilä, Sil a
Sadle , D. E., Wa s, P. C., & Uusi-Heikkilä, S. (2023). The Riddle o How Fishe ies In luence
Gene ic Di e si y. Fishes, 8(10), A icle 510. h ps://doi.o g/10.3390/ ishes8100510
2023
Ci a ion: Sadle , D.E.; Wa s, P.C.;
Uusi-Heikkilä, S. The Riddle o How
Fishe ies In luence Gene ic Di e si y.
Fishes 2023,8, 510. h ps://doi.o g/
10.3390/ ishes8100510
Academic Edi o : E ic Halle man
Recei ed: 31 Augus 2023
Re ised: 5 Oc obe 2023
Accep ed: 12 Oc obe 2023
Published: 13 Oc obe 2023
Copy igh : © 2023 by he au ho s.
Licensee MDPI, Basel, Swi ze land.
This a icle is an open access a icle
dis ibu ed unde he e ms and
condi ions o he C ea i e Commons
A ibu ion (CC BY) license (h ps://
c ea i ecommons.o g/licenses/by/
4.0/).
ishes
Re iew
The Riddle o How Fishe ies In luence Gene ic Di e si y
Daniel E. Sadle , Phillip C. Wa s and Sil a Uusi-Heikkilä *
Depa men o Biological and En i onmen al Science, Uni e si y o Jy äskylä, 40500 Jy äskylä, Finland;
[email p o ec ed] (D.E.S.); [email p o ec ed] (P.C.W.)
*Co espondence: [email p o ec ed]
Abs ac :
O e ishing d i es popula ion decline, which in u n d i es loss o gene ic di e si y.
Many s udies p o ide e idence o declines in gene ic di e si y; howe e , con o e sy exis s wi hin
he li e a u e, as some s udies show e idence o no change in gene ic di e si y despi e decades
o o e ha es ing. The appa en disc epancy in he li e a u e should he e o e be examined o
unde s and wha biological and ecological p ocesses a e d i ing he di e ences in esul s. He e,
we assess how di e en ac o s con ibu e o ishe ies-induced suscep ibili y o declines in gene ic
di e si y by i s ocusing on he di e en oles o gene ic ma ke s. Second, we assess how habi a
ype and condi ions con ibu e o loss o gene ic di e si y. Thi d, we assess how li e his o y and
physiology a ec s ca chabili y and loss o gene ic di e si y. Finally, we discuss how coinciding abio ic
and bio ic ac o s in luence he in ensi y o gene ic loss. We ind a mul i ude o hese ac o s could be
in e ac ing o in luence how esul s a e pe cei ed and how in ense he loss o gene ic di e si y can
be. Fu u e s udies should ca e ully conside he me hodology o gene ic analysis used, as well as
conside a ions o li e his o y and ecology o he a ge species.
Keywo ds: gene ic di e si y; ishe ies; conse a ion gene ics; size selec ion; popula ion gene ics
Key Con ibu ion:
This e iew assesses he impo ance o conside ing mul iple ac o s ha could
a ec he ulne abili y o an o e ished popula ion o loss o gene ic di e si y.
1. In oduc ion
Fish s ocks a e a c ucial sou ce o li elihood and nu i ion, p o iding 156 million
onnes o ood pe annum, ep esen ing some 17% o global animal p o ein consumed [
1
].
Howe e , because o a 122% inc ease in ish consump ion along wi h his o ic o e ha es -
ing, he pe cen age o ish s ocks ha a e ha es ed sus ainably has educed om 90% o
65.8% om 1990 o 2017 [
1
]. O e ishing no only causes conce ns o global ood secu i y,
bu also emains a he o e on o he an h opogenic p oblems he aqua ic en i onmen
aces, causing se e e popula ion declines and habi a des uc ion [
2
–
4
]. Indeed, he mode n
consensus is ha despi e he high ecundi y o many ish species (e.g., wi h indi iduals o
ce ain species capable o p oducing ens o hund eds o housands o eggs in a spawning
e en ), his is no enough o main ain a sus ainable s ock unde in ensi e ha es ing [
5
].
Cu en p edic ions sugges ish s ocks may become exhaus ed by 2050 [
6
,
7
], which no only
esul s in popula ion losses, bu a po en ial loss o s anding gene ic di e si y [
8
], leading o
localised ecosys em collapse, as al eady expe ienced in places such as he Black Sea [9].
Because ishing mo ali y can be ex emely high and ishe ies a e o en also selec i ely
emo ing ce ain pheno ypes om he popula ion (i.e., exposing ish popula ions o di-
ec ional selec ion), ishe ies can be expec ed o educe gene ic di e si y in an exploi ed
popula ion. Indeed, di ec ional selec ion, o en occu ing h ough size-selec ion a ou ing
he smalles indi iduals in a popula ion, will no only ac on pheno ypic ai s (e.g., body
size, age a ma u a ion, and ep oduc i e ou pu [
10
]), bu also on he associa ed gene ic
s uc u e. As such, di ec ional selec ion can ac in andem wi h demog aphic loss o en-
hance he loss o gene ic di e si y. Decline in gene ic di e si y can lead o educed adap i e
Fishes 2023,8, 510. h ps://doi.o g/10.3390/ ishes8100510 h ps://www.mdpi.com/jou nal/ ishes
Fishes 2023,8, 510 2 o 14
po en ial [
11
,
12
]. Howe e , he e is some deba e whe he neu al di e si y can be uly
ep esen a i e o adap i e po en ial, as he genes de ec ed do no ha e any di ec e ec on
popula ion i ness [
13
]. Reduc ions in adap i e po en ial can lead o inc eased ulne abili y
o u u e an h opogenic s esso s. The educ ion in gene ic di e si y can cause educed
pheno ypic a iabili y, which in u n may a ec popula ion iabili y and s abili y [
14
].
Indeed, e idence o educed gene ic di e si y in exploi ed ish s ocks has been epo ed
by a ious s udies, o example, as an inc ease in inb eeding coe icien o a educ ion in
e ec i e popula ion size [8,15,16].
Many s udies ha e ied o assess he e ec o ishing s ess on he gene ic di e si y o
ish popula ions wi h mixed esul s, wi h some s udies showing clea declines in gene ic
di e si y. Fo example, Hause e al. [
15
] showed ha he o e exploi a ion o he New
Zealand snappe (Pag us au a us) esul ed in a la ge educ ion in bo h gene ic di e si y
(he e ozygosi y and numbe o alleles) and e ec i e popula ion size (N
e
). Howe e , Jones
e al. [
17
] ound highe N
e
alues han expec ed in he o iginal s udy by inco po a ing a
la ge sample size and simula ion-based me hods, hough, despi e his, N
e
was s ill ound
o be lowe in exploi ed popula ions han in non-exploi ed popula ions. In con as , o he
s udies ha e ound no signi ican educ ions in gene ic di e si y in o e exploi ed s ocks,
such as in A lan ic cod (Gadus mo hua) s ocks [
5
,
18
–
20
]. Howe e , s ocks expe iencing
ela i ely low ca ch a es, such as mango ilapia (Sa o he odon galilaeus) in he sea o Galilee,
show a signi ican loss in gene ic di e si y [
21
]. Pinsky and Palumbi [
8
] comple ed a
comp ehensi e me a-analysis u ilising 140 species, demons a ing an o e all end ha
o e ha es esul ed in a decline in gene ic di e si y ac oss ma ine ish species, while
some s udies show no appa en gene ic loss when exposed o ishing s ess [
16
,
22
,
23
].
I is he e o e impo an o unde s and why some s udies appa en ly demons a e clea
e idence o loss o gene ic di e si y when exposed o ishing s ess, whils o he s do no .
The possible easons why some s ocks do no exhibi a no able loss o gene ic di e si y
appea complex, and he li e a u e emains di ided.
The magni ude o mo ali y a es and he in ensi y o di ec ional selec ion de e mine
how much ishe ies can e ode gene ic di e si y in a popula ion, alongside ini ial popula ion
size, e ugia om ishe ies, and mig a ion a e [
8
]. Howe e , he a iance in esul s om
he li e a u e sugges s ha many o he ac o s ha in luence gene ic di e si y mus be
conside ed. This disc epancy may be a esul o changes in gene ic echnologies esul ing
in di e ing esolu ions be ween gene ic ma ke s such as single-nucleo ide polymo phisms
(SNPs) and mic osa elli es (e.g., [
24
]). Fu he mo e, habi a di e si y may ha e an in luence,
o example, due o ac o s such as habi a complexi y ha may a ec ulne abili y o being
caugh by ishing gea (i.e., ca chabili y; [
25
,
26
]), species me apopula ion s uc u e (which
de e mines pa e ns o dispe sal among popula ions), and/o whe he a species’ ange
encompasses p o ec ed a eas. Mo eo e , li e his o y and beha iou al ai s can a ec
ca chabili y and consequen ly loss o gene ic di e si y [
27
]. I is also impo an o no e he
e olu iona y his o y o a popula ion, and how di e en species di e in hei ulne abili y
o ishe ies s ess based on hei e ol ed se o li e his o y ai s [
28
]. Finally, o e ishing
is no he only s esso e iden in he aqua ic en i onmen , wi h u he an h opogenic
e ec s such as clima e change and habi a loss likely con ibu ing o exace ba ing he loss
o gene ic di e si y [25].
In his li e a u e e iew, we will assess how di e en ac o s con ibu e o ishe ies-
induced suscep ibili y o educ ions in gene ic di e si y and aim o un a el why disc ep-
ancy in popula ion esponses in he li e a u e exis s. Fi s , we ocus on he ole o di e en
gene ic ma ke s and me ics used o measu e a iabili y in gene ic di e si y p oduced by
he di e en s udies. Second, we conside how habi a ype and condi ions can con ibu e
o he ishe ies-induced loss o gene ic di e si y. Thi d, we assess he e ec o li e his o y
ai s on ca chabili y and loss o gene ic di e si y, and inally discuss how coinciding ac o s
in luence he in ensi y o gene ic loss.
Fishes 2023,8, 510 3 o 14
2. Gene ic Ma ke s and Me ics
S udies ocusing on de ec ing po en ial ishe ies-induced changes in gene ic di e si y
ha e used di e en gene ic ma ke s. One eason why he in e p e a ions o ishe ies’ e ec s
on gene ic di e si y a y could be ela ed o he ma ke s oge he wi h he me ics u ilised
in hese s udies [
24
,
29
]. He e, we e iew he mos common ma ke s used in he s udies
and discuss how and why hey migh p oduce di e en esul s.
2.1. AFLPs
Pionee ing s udies on gene ic di e si y u ilised DNA inge p in ing in he o m o
ampli ied agmen leng h polymo phisms (AFLPs). This echnique u ilises es ic ion
enzymes o diges DNA, c ea ing es ic ion agmen s which a e hen ampli ied, allowing
he de ec ion o DNA polymo phisms [
30
]. AFLPs we e popula as hey we e an easy
and cheap way o ob ain high-quali y genomic da a [
31
,
32
], and hey ha e been used in
gene ic di e si y s udies in ishe ies esea ch. Fo example, Wan e al. [
33
], used DNA
inge p in ing o assess changes in he gene ic di e si y o Dab y’s s u geon (Acipense
dab yanus) om 1958 o 1999 a e la ge-scale declines, pa ly due o o e ishing, and
showed a signi ican educ ion in gene ic di e si y. AFLPs a e obus , highly ep oduceable,
and can be in o ma i e in la ge numbe s [
31
]. Howe e , AFLPs a e o en di icul o analyse,
equi ing high-quali y, highly pu e DNA, and as a dominan ma ke hey do no de ec
he e ozygo es. Hence, he echnique has been mos ly eplaced by o he echniques [
31
,
34
].
2.2. Mi ochond ial DNA
Mi ochond ial DNA (m DNA) is being used mo e o en o assess changes in gene ic
di e si y compa ed o AFLPs [
21
,
35
–
38
]. This echnique u ilises cy och ome c oxidase
subuni I (COI) and he D-loop sec ion o m DNA. In e es ingly, s udies using m DNA
ubiqui ously show a decline in gene ic di e si y caused by ishe ies [
37
–
39
]. Fo example,
Johnson e al. [
37
] (2018) showed ha di e si y signi ican ly dec eased a e his o ical ha -
es ing o Chinook salmon (Onco hynchus shawy scha). Mo eo e , in a s udy on swo d ish
(Xiphias gladius) in he Medi e anean, a loss o haplo ype di e si y and N
e
was de ec ed in
emales, which a e la ge and po en ially mo e hea ily a ge ed by ishe ies [
38
]. m DNA may
be mo e sensi i e o gene ic loss han genomic DNA; howe e , m DNA exhibi s a much slowe
a e o e olu ion han genomic DNA due o i s highly conse a i e
na u e [40–42]
. The e o e,
m DNA is conside ed less use ul o examining ecen changes in gene ic di e si y. Howe e ,
when assessing his o ical DNA alongside genomic DNA, m DNA is a sui able choice o
ma ke . Fo example, B o oski e al. [
21
] used a combina ion o mic osa elli es and m DNA o
show ha mango ilapia expe ienced a bo leneck e en dec easing gene ic di e si y, possibly
due o a combina ion o o e ishing and decline in en i onmen al condi ions.
2.3. Mic osa elli es
Mic osa elli es emain he gene ic ma ke o choice in he majo i y o s udies analysing
gene ic di e si y in exploi ed popula ions [
43
–
45
]. This is due o hei inhe en high
polymo phism, apid mu a ion a e, and ela i ely low cos [
43
,
46
]. Ea ly s udies on gene ic
di e si y ypically used a small numbe o mic osa elli es, e.g., [
47
,
48
]. This may ha e
masked disc e e changes in di e si y, as i has been demons a ed ha s udies conduc ed
wi h ewe han 10 mic osa elli e loci ha e low s a is ical powe o de ec changes in gene ic
di e si y [
8
,
49
–
51
]. Only a mino i y o s udies ha e used mo e han 10 loci o de ec changes
a e an o e ha es ing e en ; hese include a lake ou (Sal elinus namaycush) s udy wi h
18 loci [
52
], a s udy on dusky kob (A gy osomus japanicus) wi h 18 loci [
53
], and a s udy on
Eu opean eel (Anguilla anguilla) u ilizing 22 loci [54].
Despi e being a popula choice o gene ic ma ke , mic osa elli es ha e one majo
disad an age. They a e neu al ma ke s, meaning hey occu in non-coding egions in he
genome as opposed o o he echniques like SNPs ha occu in coding egions. The e o e,
whils hey can be used o es ima e changes in, o example, e ec i e popula ion size
(N
e
), hey canno de ec loss o adap i e gene ic di e si y as unc ional ma ke s can [
55
].
Fishes 2023,8, 510 4 o 14
Ins ead o using mic osa elli es alone, a mo e op imal solu ion could be a combina ion o
mic osa elli es, SNPs, and educed ep esen a ion lib a y (RRL) sequencing [24,29].
2.4. SNPs
SNPs a e cu en ly he eme ging gene ic ma ke o s udies on ishe ies’ e ec s on
gene ic di e si y [
56
–
58
]. SNPs a e a po en ial eplacemen o mic osa elli es as hey o e
g ea e powe o de ec ion o gene ic change han small numbe s o mic osa elli e ma k-
e s [
59
]. SNPs can be u ilised in a a ie y o ways om using a SNP a ay ac oss he genome
(e.g., [
10
]) o sequencing me hods using a la ge numbe o SNPs such as in Res ic ion
Si e Associa ed sequencing (RADseq; e.g., [
60
]), RNA sequencing [
61
], and whole-genome
sequencing [
58
]. Fu he mo e, SNPs can be used as diagnos ic ma ke s, which can explain
a la ge amoun o a ia ion in one ai . Ba son e al. [
62
] used 200,000 SNPs and ound
a la ge e ec locus (VGLL3) esponsible o age a ma u a ion in A lan ic salmon (Salmo
sala ). They ocused on his pa icula locus and showed ha i was esponsible o 39% o
he pheno ypic a ia ion obse ed in he popula ion; changes in he allele equency o his
loci can help de ec e olu iona y changes in age and size a ma u a ion in exploi ed salmon
popula ions, in o ming managemen s a egies.
D’Aloia e al. [
24
] compa ed h ee gene ic ma ke s, namely mic osa elli es, non-
epe i i e nuclea loci, and SNPs, o de e mine how each ma ke ype could al e he
gene ic me ics p oduced in a popula ion gene ics s udy. They sequenced he DNA o a
ee ish (Elacan inus lo i) using 2418 SNPs, 89 mic osa elli es, and 57 non- epe i i e nuclea
loci. The SNPs showed highe F
ST
alues ( he amoun o gene ic a iance wi hin a sub-
popula ion compa ed o gene ic a iance wi hin he en i e popula ion; [
63
]) han he o he
ma ke s and showed g ea e esolu ion o gene ic s uc u e, allowing o mo e ine-scaled
analysis. Howe e , when compa ing o he di e si y me ics, such as expec ed he e ozygos-
i y, obse ed he e ozygosi y, and allele ichness, alues we e highes in he mic osa elli e
da a. I is also impo an o no e ha mic osa elli e and non- epe i i e nuclea loci a e
cheape op ions, and a la ge numbe o hem may be mo e cos -e ec i e han SNPs [
29
,
60
].
Despi e his, mode n sequencing me hods u ilizing a high numbe o SNPs may sol e some
o he p io men ioned p oblems, as shown by mo e ecen s udies [
23
,
56
–
58
,
64
]. Ye , mixed
esul s s ill p eside wi hin hose s udies u ilising SNPs, which may be explained by he low
numbe o SNPs used. The kildsen e al. [
58
] u ilised o e wo million SNPs and showed a
signi ican educ ion in gene ic di e si y (nucleo ide di e si y and polymo phism %) in an
expe imen ally ha es ed ish popula ion. In con as , Pinsky e al. [
65
] used app oxima ely
350,000 SNPs and demons a ed no change in gene ic di e si y (genome-wide di e si y)
despi e la ge-scale ha es ing o cod. Sadle e al. (Unpublished), on he o he hand,
u ilized almos 3.5 million SNPs and showed a dec ease in gene ic di e si y (he e ozygosi y,
nucleo ide di e si y, and N
e
) in size-selec i ely ha es ed expe imen al zeb a ish (Danio
e io) popula ions.
When di e en gene ic ma ke s a e u ilised, di e en me ics a e epo ed, and hese
a e no always compa able. Fo example, allele ichness and di e si y a e mo e p one o
educ ions in popula ion size han he e ozygosi y [
14
,
66
]. The e o e, i a s udy only u ilises
he e ozygosi y as a me ic (e.g., [
48
]), o he aspec s o gene ic di e si y change could be
o e looked. C ucially, s udies should conside ca e ully he implica ions o me hodology
used, aiming o he mos e icien , low-cos me hod ha a ge s adap i e loci.
3. Habi a Type
3.1. F eshwa e s. Sal wa e
Habi a ype has a signi ican e ec on he ulne abili y o aqua ic species o ex-
ploi a ion, he mos ob ious di ide being he di e ence be ween eshwa e and sal wa e
habi a s, oge he wi h he in e media e b ackish en i onmen . Despi e he la ge di e ence
in habi a size o global aqua ic en i onmen s (0.5% accessible eshwa e and 96.5% sal -
wa e ; [
67
]) he di e si y o ish is e y simila , wi h app oxima ely 15,000 ac inop e ygian
ish species p e alen in bo h ecosys ems [
68
,
69
]. The high species di e si y obse ed in

Fishes 2023,8, 510 5 o 14
he eshwa e ealm likely o igina es om la ge deg ees o habi a agmen a ion h ough
di e en ia ed i e s and isola ed lake and pond sys ems, allowing o g ea e oppo uni y
o adap i e di e gence o ake place compa ed o he compa a i ely ubiqui ous habi a o
he oceans [68].
S uc u ally, ma ine and eshwa e sys ems a e e y di e en , wi h g ea e habi a
connec i i y in he ma ine en i onmen due o he huge expanse o a ailable habi a space.
Indeed, gene ically, ma ine ish exhibi less gene ic s uc u e han eshwa e species [
20
,
70
].
I may be p esumed ha eshwa e habi a s a e mo e p one o gene ic loss due o he e
being ewe e ugia om ishe ies and lowe connec i i y es ic ing mig a ion. Despi e
his, mos s udies ocus on ma ine species due o hei inhe en comme cial impo ance,
wi h a cap u e o 84 million onnes compa ed o 12 million onnes in eshwa e ishe ies in
2018 [
1
]. Howe e , eshwa e en i onmen s expe ience high ishing p essu e [
71
], ye less
is known abou how ishe ies a ec hei gene ic di e si y. S udies ha ocus on eshwa e
sys ems, show a g ea e decline in gene ic di e si y compa ed wi h s udies ocusing on
ma ine habi a s (Figu e 1a; Table S1). F eshwa e sys ems may he e o e be mo e p one o
loss o gene ic di e si y when exposed o ishe ies s ess.
Fishes 2023, 8, 510 5 o 15
di e ence in habi a size o global aqua ic en i onmen s (0.5% accessible eshwa e and
96.5% sal wa e ; [67]) he di e si y o ish is e y simila , wi h app oxima ely 15,000 ac-
inop e ygian ish species p e alen in bo h ecosys ems [68,69]. The high species di e si y
obse ed in he eshwa e ealm likely o igina es om la ge deg ees o habi a agmen-
a ion h ough di e en ia ed i e s and isola ed lake and pond sys ems, allowing o
g ea e oppo uni y o adap i e di e gence o ake place compa ed o he compa a i ely
ubiqui ous habi a o he oceans [68].
S uc u ally, ma ine and eshwa e sys ems a e e y di e en , wi h g ea e habi a
connec i i y in he ma ine en i onmen due o he huge expanse o a ailable habi a
space. Indeed, gene ically, ma ine ish exhibi less gene ic s uc u e han eshwa e spe-
cies [20,70]. I may be p esumed ha eshwa e habi a s a e mo e p one o gene ic loss
due o he e being ewe e ugia om ishe ies and lowe connec i i y es ic ing mig a-
ion. Despi e his, mos s udies ocus on ma ine species due o hei inhe en comme cial
impo ance, wi h a cap u e o 84 million onnes compa ed o 12 million onnes in esh-
wa e ishe ies in 2018 [1]. Howe e , eshwa e en i onmen s expe ience high ishing
p essu e [71], ye less is known abou how ishe ies a ec hei gene ic di e si y. S udies
ha ocus on eshwa e sys ems, show a g ea e decline in gene ic di e si y compa ed
wi h s udies ocusing on ma ine habi a s (Figu e 1a; Table S1). F eshwa e sys ems may
he e o e be mo e p one o loss o gene ic di e si y when exposed o ishe ies s ess.
Figu e 1. Change in he e ozygosi y associa ed wi h ha es ing (a) ish inhabi ing ma ine and esh-
wa e (FW) habi a s, (b) o ganized by o de s, and li e his o y ai s, such as (c) pa i y and (d) mig a-
ion s a egy (see Table S1 o da a sou ces). A nega i e alue indica es loss o he e ozygosi y a e
a ha es ing e en . Dashed line indica es no change in he e ozygosi y. In panels (a,c), shaded a eas
ep esen he dis ibu ion o he da a as densi y, and in panels (b,d), da a a e ep esen ed as means
and s anda d e o s.
Fu he mo e, in ma ine sys ems, he di e ing occu ence o ish popula ions can in-
luence hei ca chabili y, dependen on he ishing me hod used, subsequen ly
Figu e 1.
Change in he e ozygosi y associa ed wi h ha es ing (
a
) ish inhabi ing ma ine and eshwa-
e (FW) habi a s, (
b
) o ganized by o de s, and li e his o y ai s, such as (
c
) pa i y and (
d
) mig a ion
s a egy (see Table S1 o da a sou ces). A nega i e alue indica es loss o he e ozygosi y a e a
ha es ing e en . Dashed line indica es no change in he e ozygosi y. In panels (
a
,
c
), shaded a eas
ep esen he dis ibu ion o he da a as densi y, and in panels (
b
,
d
), da a a e ep esen ed as means
and s anda d e o s.
Fu he mo e, in ma ine sys ems, he di e ing occu ence o ish popula ions can in lu-
ence hei ca chabili y, dependen on he ishing me hod used, subsequen ly in luencing
ulne abili y o gene ic loss. Fo example, a awle ishe y will likely ha e a g ea e
impac on ben hic ish, whils long lines will a ge pelagic ish. In e es ingly, mo e s udies
show ha ben hopelagic ish a e mos ulne able o declines in gene ic di e si y ollowing
Fishes 2023,8, 510 6 o 14
ishe ies s ess (Table S1), po en ially because hey li e close o he bo om o he ocean
whe e ishing me hods such as awling can cause signi ican damage o hei habi a and
cause di ec mo ali y [72].
3.2. La i ude
Species di e si y, unc ional di e si y and gene ic di e si y dec ease wi h la i ude
because o pos glacial coloniza ion [
73
,
74
]. This c ea es a g adien in species numbe , such
ha species di e si y is g ea es a he equa o [
75
,
76
]. As a esul , ishing p essu e could
be expec ed o be mo e in ense a he equa o due o a high numbe o ishing communi ies
and species. Howe e , his is no he case, as ishing ho spo s in Eu ope, No h Ame ica,
and China skew his pa e n. Because no he n (and pola ) popula ions demons a e lowe
le els o gene ic di e si y han sou he n ones, high ishing p essu e in hese a eas migh
ha e g ea e impac on gene ic di e si y han an icipa ed [
77
]. Shi ing baseline e e s
o he phenomenon o how people’s pe cep ions, o example, o gene ic di e si y (o
biodi e si y), change o e ime based on hei own expe ience, memo y, o knowledge
(e.g., [
78
]). Fo example, people may no ealise how much gene ic di e si y has been los
in ce ain species o popula ions because hey ha e no e e ence poin o compa e wi h he
pas [
79
]. This means we should ca e ully conside he imescale o ishe ies, and a wha
poin s loss o indeed no loss o gene ic di e si y is obse ed. This can esul in a lack o
awa eness o conce n o conse a ion and ishe ies managemen [80].
3.3. Habi a Complexi y
Habi a ange and complexi y a e impo an ac o s in luencing a ish popula ion’s
ulne abili y o ishe y s ess. Isola ed, small ish popula ions a e mo e p one o he
loss o gene ic di e si y han la ge, well- connec ed popula ions [
81
]. Connec i i y o a
habi a is c ucial in main aining gene ic di e si y, as i allows o gene ic eco e y h ough
mig a ion [
8
]. Fu he mo e, a anging, panmic ic species a e mo e likely o ha e si es o
e ugia om ishe ies, allowing o a eas o high gene ic di e si y ha can hen con ibu e
o he o e all popula ion h ough mig a ion. Mo eo e , physical ba ie s can inhibi
connec i i y o popula ions, pa icula ly in eshwa e en i onmen s whe e lakes and i e s
can become cu o . In ole ance o ce ain en i onmen al condi ions such as salini y, oxygen,
and empe a u e can also c ea e ba ie s o connec i i y. Fo example, Vi o ino e al. [
82
]
s a e ha A apaima gigas li es in len ic wa e s, and a e in ole an o lo ic en i onmen s ha
ac as ba ie s o dispe sal, which may inc ease his species’ suscep ibili y o loss o gene ic
di e si y due o lack o e ugia and mig a ion oppo uni y.
When p edic ing he po en ial loss o gene ic di e si y induced by ishing, a ious
ac o s o he han ishing in ensi y, selec i i y, and ype o gene ic ools used, con ibu e.
Unde wa e habi a ype, condi ion, and complexi y can be di icul o always de e mine
accu a ely. Howe e , s udies sugges ha hese can possibly magni y he a e a which
ishe ies can deg ade gene ic di e si y and, he e o e, hey should be conside ed.
4. Popula ion Demog aphy
4.1. Popula ion Size and Range
La ge popula ions a e usually less ulne able o andom gene ic d i and he e o e
gene ic di e si y is less likely o be deg aded unde in ense ishing s ess. Indeed, some
ha e e en sugges ed ish s ocks a e so la ge ha e en collapsed popula ions a e comple ely
esis an o loss o gene ic a ia ion h ough andom gene ic d i [
83
]. Howe e , e ec i e
popula ion size (N
e
; i.e., he numbe o indi iduals ha de e mines gene ic cha ac e is ics,
such as allele equency) is o en signi ican ly smalle han he ac ual numbe o ish in a
popula ion (census popula ion size, N; [
84
]). I was he e o e o en assumed ha gene ic
bo lenecks we e a e in la ge ish popula ions when conside ing census popula ion size;
howe e , gene ic d i is de e mined by he e ec i e popula ion size ha is many ac o s
lowe [
15
]. Some au ho s ha e e en sugges ed ha educ ion in allele di e si y as a esul
o o e ishing may be mo e in ense in la ge popula ions han small ones [
85
]. When a
Fishes 2023,8, 510 7 o 14
popula ion is ha es ed down o low numbe s, i s inb eeding coe icien inc eases (i.e.,
gene ic di e si y dec eases), inb eeding dep ession occu s, i ness is educed oge he wi h
he popula ion’s e olu iona y po en ial, and he popula ion can e en ually be des ined o
ex inc ion [86].
In addi ion o popula ion size, i s ange can u he de e mine ulne abili y o ishe ies
exploi a ion. A widely dis ibu ed species such as una (Thunnus hynnus) is mo e likely o
be esis an o gene ic loss caused by ishe ies due o po en ial e ugia in he en i onmen
and g ea e habi a connec i i y [
23
]. Con e sely, a na owly dis ibu ed species may be
mo e p one o gene ic loss because gene ic di e si y canno be main ained by e ugia u i-
liza ion and ebuil h ough mig a ion [
8
]. Fo example, Maugean ska e (Zea aja maugeana)
is endemic o Tasmania, has a na ow ange, is endange ed, and has su e ed om he loss
o gene ic di e si y, likely due o a combina ion o ishing and en i onmen al change [87].
4.2. Sex Ra io
Fishing egimes o en a ge he la ge ish in a popula ion, which no only in luences
body size and ep oduc ion di ec ly, bu also leads o sex a io al e a ion in sexually
dimo phic species [
38
,
88
]. In ac , “sex-selec i e” ishe ies can exace ba e he loss o gene ic
di e si y by hampe ing ep oduc ion di ec ly by making he o he sex a e, hence lowe ing
he N
e
[
89
]. Fo example, in he Gul o Mexico, popula ions o ambe jack (Se iola dume ili)
ha e a emale-skewed sex a io, which is likely due o ca ch p e e ences in he egion:
emales a e a mo e alued ca ch because hey a e signi ican ly la ge han males, making
o a mo e p o i able ca ch [
90
]. Fu he mo e, Righi e al. [
38
] showed ha unde ishing
s ess, male swo d ish (Xiphias gladius) became mo e p e alen in he popula ion as a esul
o ishe men a ge ing he la ge emales, leading o losses in gene ic di e si y. A skewed
sex a io is a p oblem in all sexually dimo phic ish species ha a e being exploi ed, as
he la ge sex is mo e likely o be cap u ed. This can lead o gene ic loss and e en o
ex inc ion e en s i he ca ch a e is ex eme. To p e en ishe ies o lead o skewed sex a io,
a maximum size limi could be help ul in p o ec ing he po en ially la ge sex ( ypically
emale). Howe e , e en mo e s ingen managemen measu es, such as pe manen ishe ies
closu e can be e ec i e only when implemen ed in he long e m. A demog aphic eco e y
o a ma ine ee ish ook 30 yea s o seasonal ishe ies closu e [
91
]. A gene ic eco e y
likely akes much longe .
Fu he mo e, es ima ing popula ion sizes (census o e ec i e) o o iginal s anding
gene ic di e si y can be challenging. Ye , hese wo popula ion-le el cha ac e is ics a e
impe a i e i we aim o eliably p edic he a e and magni ude a which ishe ies can
induce loss o gene ic di e si y.
5. Li e His o y and Beha iou
S udies based on he o e ha es ing o popula ions a e unsu p isingly based on he
mos comme cially impo an species, such as cod (Gadus mo hua) and he ing (Clupea
ha engus). This leads o an inhe en bias on he li e his o ies and demog aphy o hese
s udies. Cod, o example, ha e massi e, in e connec ed popula ions, wi h ex emely
high li e ime ecundi y and aking a ela i ely long ime o ma u e. Ancho y and sa dine,
on he o he hand, each ma u i y apidly (1–2 yea s) bu ha e ewe spawning e en s.
Fu he con as ing a e species such as sha ks and ays ha a e o en e y long li ed, bu
wi h low ecundi y. Fish li e his o y, physiology, and beha iou a e ex emely a iable
amongs species and, in ui i ely, he e ec o ishe ies on loss o gene ic di e si y will
as ly di e . S udies on species which each a la ge body size and ha e long li e spans ha e
demons a ed a ishe ies-induced loss o gene ic di e si y, e.g., [
33
,
38
,
48
]. Howe e , he
same is e iden o small as -li ing species (e.g., [
92
,
93
]). Va ia ion exis s in he li e a u e
be ween axonomic g oups (Figu e 1b), so deepe explo a ion in o li e his o y ai s is
equi ed o unde s and pa e ns o gene ic loss.
Fishes 2023,8, 510 8 o 14
5.1. Li e His o y
In ensi y o ishe y s ess may a ec ish di e en ly depending on hei ep oduc i e
s a egy, as a pelagic ba ch spawne wi h la ae ha dispe se o e as dis ances, such as
he ing, may ha e a g ea e chance o main aining gene ic di e si y han species ha lay
eggs on o a subs a e [
94
]. The la e may be pa icula ly ulne able o awling and o he
des uc i e ishe y me hods [
95
], as ishe ies will al eady cause high mo ali ies a he egg
s age. Fu he mo e, ish ha a e b oadcas spawne s (i.e., spawning con inues o se e al
subsequen mon hs and a els a g ea e dis ance), such as una and sa dines, will likely
ha e lowe egg mo ali y a e caused by ishe ies, whils in con as , ben hic spawne s ha
lay in close p oximi y, a aching eggs o he ben hos, will be p one o he same a ge ed
ishing (e.g., cod and plaice), especially des uc i e me hods such as awling, esul ing in
highe egg mo ali y and hence an indi ec ly g ea e loss o gene ic di e si y.
Rep oduc ion in ish can be b oadly classi ied in o wo ca ego ies: semelpa ous (a
single ep oduc i e e en du ing li e ime) and i e opa ous (mul iple ep oduc i e e en s).
Semelpa ous ish a e hough o be mo e p one o loss o gene ic di e si y as hey ely on a
singula ep oduc i e e en , so ep oduc i e ou pu is dependen on he a ailable pa ne a
he ime o ep oduc ion, as well as o sp ing su i al being dependen on he empo ali y
o ep oduc ion [
96
]. Fo example, ep oduc ion may coincide wi h an en i onmen al
s esso o indeed an in ense ishe y e en . In con as , i e opa ous ish sp ead hei
ep oduc i e ou pu ou , and a e less likely o be a ec ed by a singula en i onmen al o
ishing e en . Species, such as many salmonids, ha spawn once a yea a a speci ic loca ion
could be ex emely ulne able o ishe ies i i a ge s spawning popula ions. Fishe ies
ha e a ge ed he spawning s ocks o ba ed sand bass (Pa alab ax nebuli e ) and kelp
bass (Pa alab ax cla h a us), esul ing in la ge-scale popula ion loss [
97
]. Clea ly, ishe ies
ope a ing on spawning g ounds, pa icula ly in hose o semelpa ous ishes, can e y
e ec i ely educe popula ion sizes and he e o e gene ic di e si y. Howe e , e idence in
he li e a u e sugges s i e opa ous ish a e mo e p one o loss o gene ic di e si y (Figu e 1c),
hough his could be due o lack o s udies on semelpa ous ish, and indeed i is a a e
li e his o y s a egy. Fu he mo e, ishe y closu es du ing spawning season a e nowadays
ela i ely common and ha e been sugges ed o con ibu e o sus ainable ishing [98].
Body size has been shown o display a nega i e ela ionship wi h gene ic
di e si y [8,99].
As a esul , la ge indi iduals and la ge species a e mo e likely o ha e lowe di e si y,
due o body size and ca ch being ela ed o popula ion abundance [
99
]. This may esul
in o e ished popula ions o smalle indi iduals showing no appa en loss in gene ic
di e si y, bu which ha e los c ucial a e alleles ela ed o g ow h and i ness. Fu he mo e,
la ge ish a e usually mo e ep oduc i ely success ul in many species, p oducing mo e
and la ge eggs [
100
]. Mo eo e , la ge emales o en disp opo ionally con ibu e o
ec ui men ; he e o e, loss o he la ges indi iduals may esul in a educ ion in he
g ow h and ecundi y o he popula ion [
15
]. Howe e , con e sely, he emo al o la ge
ish ha disp opo iona ely con ibu e o he popula ion may enhance gene ic di e si y
due o he inc eased ep oduc i e success o mo e, smalle indi iduals. As such, li e his o y
s a egy as a p edic o o he magni ude o ishe ies-d i en loss in gene ic di e si y is no
s aigh o wa d and likely in e ac s wi h o he ai s and ac o s.
When examining li e his o y s a egies mos ulne able o ishing, an impo an
concep is /K-selec ion [
101
]. K-selec ed species a e la ge, ma u e la e, and p oduce
ela i ely ew o sp ing; as such, hei popula ion size, and hence gene ic di e si y, may be
mo e a ec ed by ishing, as o example demons a ed in he o ange oughy (Hoplos e hus
a lan icus; [
48
]). In con as , -selec ed species cha ac e ized by small body size, ea ly
ma u i y, and high ecundi y may be mo e du able o ishe ies s ess, as hey ha e highe
popula ion g ow h a es and may be less likely o expe ience gene ic loss. Howe e , ish
a e di icul o ca ego ise in o /K-species, as many o hem do no s ic ly ollow he
de ini ions. Fo example, cod can be desc ibed as a K-species because i has a la ge body
size and ma u es la e, a he age o i e o eigh yea s. Howe e , i is highly ecund and
can spawn mo e han 10 million eggs du ing a spawning season [
102
]. He e, we canno