Unique inhibitory cascade pattern of molars in canids contributing to their potential to evolutionary plasticity of diet.

0.59 MB·English

by  Asahara, Masakazu

Unique inhibitory cascade pattern of molars in canids
contributing to their potential to evolutionary plasticity
of diet
Masakazu Asahara
GraduateSchoolofSciences,KyotoUniversity,Kyoto,606-8502,Japan
Keywords Abstract
Canidae,Carnivora,dentalformulae,dental
Developmental origins that guide the evolution of dental morphology and den-
morphology,evolvability,inhibitorycascade.
tal formulae are fundamental subjects in mammalian evolution. In a previous
Correspondence study, a developmental model termed the inhibitory cascade model was estab-
MasakazuAsahara,GraduateSchoolof lished. This model could explain variations in relative molar sizes and loss of
Sciences,KyotoUniversity,Kyoto606-8502, the lower third molars, which sometimes reflect diet, in murine rodents and
Japan.Tel:[phone];Fax:[phone]
other mammals. Here, I investigated the pattern of relative molar sizes (inhibi-
3276;E-mail:[email protected].
tory cascade pattern) in canids, a taxon exhibiting a wide range of dietary hab-
kyoto-u.ac.jp
its. I found that interspecific variation in canid molars suggests a unique
FundingInformation inhibitory cascade pattern that differs from that in murine rodents and other
FundedbyMishimaKaiumMemorial previously reported mammals, and that this variation reflects dietary habits.
Foundation,JapanSocietyforthePromotion This unique variability in molars was also observed in individual variation in
ofScience,andMinistryofEducation, canid species. According to these observations, canid species have greater vari-
Culture,Sports,ScienceandTechnologyin
ability in the relative sizes of first molars (carnassials), which are functionally
Japan
important for dietary adaptation in the Carnivora. In conclusion, an inhibitory
cascade that differs from that in murine rodents and other mammals may have
Received:11October2012;Revised:25
October2012;Accepted:1November2012 contributed to diverse dietary patterns and to their parallel evolution in canids.
EcologyandEvolution2013;3(2):278–285
doi:10.1002/ece3.436
opmental mechanism of the evolution of dental morphol-
Introduction
ogy and dental formulae, which guide and constrain the
Dental morphology and dental formulae are important reflection of dietary adaptations (e.g. Kavanagh et al.
taxonomic traits in mammals (Ungar 2010), and are also 2007; Polly 2007; Laffont et al. 2009; Renvois(cid:1)e et al. 2009;
used for paleoecological and ecomorphological studies in Harjunmaa et al. 2012; Wilson et al. 2012). A recent
mammals because these traits reflect dietary adaptations developmental study established a developmental model
(Popowics 2003; Benton 2004; Friscia et al. 2007; Van that can explain evolution of the relative sizes of lower
Valkenburgh 2007). Patterns of adaptation are guided by molars in murine rodents (Kavanagh et al. 2007). This
the variability and evolvability of these traits (Klingenberg model, termed the inhibitory cascade model, explains the
2005; Barton et al. 2007). In fact, evolvability, that is, sys- relative sizes of the lower molars (first, second, and third
tems having variability, generating new variation (Wagner molars; M , M , and M , respectively) by the balance of
1 2 3
and Altenberg 1996; Kirschner and Gerhart 1998), is inhibitor molecules from M tooth germ and activator
1
fundamental to the evolution of traits (Klingenberg 2005; molecules from mesenchyme during dental development.
Futuyma 2010). Therefore, developmental mechanisms Inhibitor molecules inhibit the development of distal
that guide and constrain patterns of adaptation in dental molars, whereas activator molecules activate their devel-
morphology and dental formulae are crucial subjects for opment. For example, greater inhibition generates a larger
elucidating their proximate and ultimate factors, and the M and smaller M (M >> M >> M ), while lower inhi-
1 3 1 2 3
interactions between these factors during dental evolution bition and greater activation generate equal-sized molars
(Kavanagh et al. 2007; Laland et al. 2011; Wilson 2011). (M = M = M ), and moderate inhibition and activation
1 2 3
Therefore, many recent studies have focused on the devel- generate an intermediate condition (M > M > M )
1 2 3
278 ª2012TheAuthor.PublishedbyBlackwellPublishingLtd.ThisisanopenaccessarticleunderthetermsoftheCreative
CommonsAttributionLicense,which permitsuse,distribution andreproductioninany medium,providedtheoriginal
workisproperlycited.
M.Asahara UniqueInhibitoryCascadeinCanidMolars
(Kavanaghet al.2007).Thismodelcanexplaindentalvari-
Material and Methods
ations that have resulted from dietary adaptations in mur-
ine rodents (Kavanagh et al. 2007). Faunivorous murine Iexamined320specimensfrom27speciesofcanids(Cani-
species exhibit M >> M , and have lost M . Conversely, dae, Carnivora, Mammalia) (Table 1). All species were
1 2 3
herbivorous murine species have approximately equal- examined to clarify evolutionary patterns in relative molar
sizedmolars.Formostmammals,frommarsupialstovari- sizes.Thedietarypatternofeachspecieswascategorizedas
ous placental orders, the relative sizes of M , M , and M carnivorous (primarily eating mammalian flesh), omnivo-
1 2 3
change sequentially and thus were explained by the model rous (eating various foods, with neither mammalian flesh
(Polly 2007). Several authors have investigated relative nor insects comprising >50% of the diet), or insectivorous
molar sizes in several mammalian taxa and have reported (primarilyeatinginsects)usinginformationfromtheliter-
differences in the inhibitory cascade pattern between a ature (Sillero-Zubiri 2010). In order to estimate variability
number of taxa including murine rodents (Renvois(cid:1)e et al. inmolars,Iexaminedindividualvariationinrelativemolar
2009; Wilson et al. 2012). It has been noted that the vari- sizes. For this purpose, individual variation within seven
abilityofatraitinitiatestheevolvabilityofthattrait(Klin- species was examined whereby I measured >15 individuals
genberg 2005; Barton et al. 2007; Wilson 2011). It is from each species (Table 1, 2). In addition, I examined
possiblethattheuniquepatternsofinhibitorycascadethat individual variation in the presence or absence of M in
3
guide variability in a particular taxon could promote the Vulpes lagopus and Nyctereutes procyonoides to clarify
evolvability of typical molar patterns and, consequently, whether individual variation and missing of M are
3
theevolvabilityofdietinthattaxon. explained by the inhibitory cascade model. The specimens
The order Carnivora, and particularly the family Cani- ofN. procyonoidesexaminedwerethosedepositedinKyoto
dae (canids), is one of the most diverse mammalian taxa University Museum, Kyoto University, Japan, which had
in terms of dietary pattern, and exhibits parallel evolution beencollectedfromasmallisland,ChiburiIsland,Shimane
in diet (Van Valkenburgh and Koepfli 1993; Friscia et al. Prefecture, Japan. Specimens of the other species were
2007; Goswami 2010; Sillero-Zubiri 2010). Similar to fau- those deposited in the Department of Mammalogy,
nivorous murine rodents, several canid species have lost American Museum of Natural History, USA, which had
M (Sillero-Zubiri 2010); this loss is thought to be related beencollectedfromlargeareas.Imeasuredthesizeofeach
3
to the enlargement of carnassial teeth (M ) – a carnivo- molarastheprojectedareainphotostakenfromtheocclu-
1
rous adaptation for shearing flesh – and to the degenera- sal view using ImageJ software (NIH, Bethesda, MD), and
tion of molars M and M (Holliday 2010). However, compared the relative molar sizes in the morphospace: M
2 3 2
patterns of relative molar sizes and dietary adaptation in size/M size versus M size/M size (abbreviated as M /M
1 3 1 2 1
canids, and the relationship of these parameters to the vs. M /M ) (Kavanagh et al. 2007). Any given point in
3 1
inhibitory cascade model are still not clear. In this study, morphospace represents the relative sizes of the three
my primary objective was to elucidate patterns of inter- molars of a particular species or individual. I plotted the
specific variation in the relative sizes of lower molars averagevaluesforeachspeciestodescribeinterspecificvari-
in canids, and to determine the relationship of this varia- ation, and plotted each individual to describe individual
tion to the inhibitory cascade model and to dietary variation.Reducedmajoraxis(RMA)regressionswereper-
adaptations. formed on these plots after performing Anderson-Darling
Asecondobjectivewastoelucidatethevariabilitywithin normalitytest.IusedM /M asanindexofactivationver-
2 1
the species in canids that guides evolutionary patterns sus inhibition during molar development. M /M scores
2 1
(Klingenberg 2005). To achieve these objectives, I investi- between carnivorous and omnivorous species were com-
gatedindividualvariationinrelativemolarsizesasanindi- paredusingtheMann–WhitneyUtest.Further,forV. lag-
cation of variability in this parameter at the intraspecific opus and N. procyonoides, M /M scores were compared
2 1
level (Klingenberg 2005). In addition, I investigated indi- betweennormalindividualsandindividualsthatweremiss-
vidual variation in the number of teeth, as oligodonty ing M on one or both sides. Statistical analyses were per-
3
(missingteeth)isconsideredatransitionalstageintheevo- formed using Minitab 14 (Minitab, Inc., PA), and RMA
lutionofdentalformulae(e.g.Ohtaishi1986;Gianniniand regressions were performed using PAST (Hammer et al.
Simmons 2007). An earlier experimental study examining 2001). Several studies have utilized multiple regressions to
mouse development found that the number of molars was elucidatehowabsolutemolarsizesaffectoneanother(Ren-
affected by the inhibitory cascade (Kavanagh et al. 2007). vois(cid:1)e et al. 2009; Wilson et al. 2012). However, this
Therefore, I also compared individual variation in rela- method tends to reflect variability in the absolute size of
tive molar sizes and number of molars (i.e. congenital M , and activation versus inhibition patterns can become
1
missing of M ), to consider the inhibitory cascade and the obscured.Therefore, I focusedonrelative molar sizes,that
3
evolutionaryprocessofM lossincanids. is,theinhibitorycascade.
3
ª2012TheAuthor.PublishedbyBlackwellPublishingLtd. 279
UniqueInhibitoryCascadeinCanidMolars M.Asahara
Table1. Speciesexaminedinthisstudy,andtheirmolarratios.
Number Species Diet N M2/M1(cid:1)SD M3/M1(cid:1)SD
1 Atelocynusmicrotis Omnivorous 3 0.54(cid:1)0.04 0.15(cid:1)0.04
2 Canisadustus Omnivorous 4 0.51(cid:1)0.12 0.17(cid:1)0.05
3 Canisaureus Omnivorous 11 0.42(cid:1)0.04 0.11(cid:1)0.02
4 Canislatrans Carnivorous 51 0.37(cid:1)0.03 0.09(cid:1)0.02
5 Canislupus Carnivorous 28 0.31(cid:1)0.03 0.09(cid:1)0.01
6 Canismesomelas Omnivorous 20 0.36(cid:1)0.02 0.10(cid:1)0.02
7 Cerdocyonthous Omnivorous 5 0.54(cid:1)0.06 0.16(cid:1)0.03
8 Chrysocyonbrachyurus Omnivorous 3 0.46(cid:1)0.01 0.18(cid:1)0.01
9 Cuonalpinus Carnivorous 2 0.25(cid:1)0.00 0.00(cid:1)0.00
10 Lycalopexculpaeus Omnivorous 10 0.41(cid:1)0.03 0.11(cid:1)0.02
11 Lycalopexgriseus Omnivorous 15 0.50(cid:1)0.05 0.13(cid:1)0.02
12 Lycalopexgymnocercus Omnivorous 9 0.53(cid:1)0.02 0.13(cid:1)0.01
13 Lycalopexsechurae Omnivorous 5 0.53(cid:1)0.04 0.16(cid:1)0.01
14 Lycalopexvetulus Insectivorous 6 0.71(cid:1)0.12 0.25(cid:1)0.11
15 Lycaonpictus Carnivorous 7 0.32(cid:1)0.02 0.07(cid:1)0.02
16 Nyctereutesprocyonoides Omnivorous 44 0.47(cid:1)0.03 0.07(cid:1)0.04
17 Otocyonmegalotis Insectivorous 7 0.97(cid:1)0.05 0.82(cid:1)0.06
18 Speothosvenaticus Carnivorous 4 0.17(cid:1)0.04 0.00(cid:1)0.00
19 Urocyoncinereoargenteus Omnivorous 31 0.53(cid:1)0.03 0.14(cid:1)0.03
20 Vulpesbengalensis Omnivorous 2 0.61(cid:1)0.03 0.23(cid:1)0.04
21 Vulpeschama Omnivorous 1 0.62(cid:1)0.00 0.20(cid:1)0.00
22 Vulpeslagopus Carnivorous 31 0.32(cid:1)0.04 0.07(cid:1)0.04
23 Vulpesmacrotis Omnivorous 4 0.39(cid:1)0.05 0.10(cid:1)0.00
24 Vulpespallida Omnivorous 1 0.72(cid:1)0.00 0.23(cid:1)0.00
25 Vulpesvelox Omnivorous 7 0.39(cid:1)0.01 0.09(cid:1)0.01
26 Vulpesvulpes Carnivorous 3 0.35(cid:1)0.01 0.10(cid:1)0.00
27 Vulpeszerda Omnivorous 6 0.58(cid:1)0.04 0.16(cid:1)0.03
canids varied greatly in relation to that of murine
Results
rodents. Consequently, the maximum relative size of M
1
As a result of interspecific variation, plots of the molar occupied 80% of the total molar row in canids, but only
ratios of all species in morphospace indicated that relative 66% of that in murine rodents (Fig. 1). Carnivorous spe-
molar sizes changed sequentially (i.e. M > M > M ; cies tended to have lower M /M scores than omnivorous
1 2 3 2 1
plots are in the white zone in Fig. 1). These results are in species (two-sided U test, W = 21.0, P < 0.001). That is,
agreement with the consensus area of the inhibitory some carnivorous species have much larger M and smal-
1
cascade model suggested by Polly (2007). Interspecific ler M and M , than omnivorous species. Notably, species
2 3
variation in relative molar sizes among the majority of at the extremes of the distribution, that is, Cuon alpinus
canid species, excluding Otocyon megalotis, exhibited a and Speothos venaticus, have lost M . These differences in
3
pattern that differed in slope from the variation observed relative molar sizes and loss of M have evolved in paral-
3
in murine rodents (Kavanagh et al. 2007). That is, the lel within many clades of canids (Fig. 3).
area in which O. megalotis was plotted indicated that it As the result of individual variation, normality of the
had similar sized molars (Fig. 1). Including O. megalotis M /M data was not observed for V. lagopus and
3 1
data in M /M scores caused the assumption of normality N. procyonoides, species in which individuals were missing
3 1
to be violated; therefore, this species was excluded from M . When individuals with dental anomalies were
3
interspecific regression analysis. The pattern of interspe- excluded, the data for all species were normally distrib-
cific variation revealed a correlation between M /M and uted (P < 0.05), and regression analyses were performed.
2 1
M /M (P < 0.001; Table 2). The slope of the variation In the M /M versus M /M morphospace, individual
3 1 2 1 3 1
among murine rodents was higher than, and outside of variations were correlated in most of the species
the 95% confidence interval of, that of canids (Table 2), (P < 0.05; Table 2), with the exception of Canis mesom-
indicating that the cascade patterns of canids and murine elas (P = 0.06; Table 2). That is, individuals with rela-
rodents are significantly different (Fig. 1, Table 2). tively larger M tended to have relatively smaller M and
1 2
According to this pattern, the relative sizes of M in M , and vice versa (Fig. 2). Ten individuals of V. lagopus
1 3
280 ª2012TheAuthor.PublishedbyBlackwellPublishingLtd.
M.Asahara UniqueInhibitoryCascadeinCanidMolars
0.9 (a) O. megalotis and 25 individuals of N. procyonoides were missing M3
0.8 17 on one or both sides (32% and 56%, respectively). There
1 000...765 Murine rodents wtoepaeustshnaowndeerveNid.ceonpnrcoescidyooefnrecodoidnetcsor,esibnceednicvcoiedn,ugaaenlnsditmaalli.lsscIinansgebsooonthfemoVri.ssltiawngog-
M
M3/ 00..43 Canids tMha3nhandorlmowalerinscdoivriedsufaolsr M(o2n/eM-s1id(eid.e.Ugretaetsetr, iWnh=ibi1ti6o9n.0)
20 14
000...012 18 922155 26642150233186112113211792721 24 mawniitdshsin5rg0el9aM.t0i3,v.elryesplaercgteivrelMy,1Pan<d0s.0m5a)lle(rFiMg.22)t.enIdneddivitdouables
0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0
Discussion
M2/M1
0.9 (b) The inhibitory cascade pattern in canids and
0.8 its relationship to diet
3/M1 0000....7654 Murine rodents Hervivorous SpMeloq2t/usMean1rteiaainlndchthMaen3gw/eMhsiit1nehrzeaolvnaeteivbieneemFniogcl.aor1n)ssiaizdneesdr(eci.doer.arMesl1aetv>ioidnMenb2cee>twtMehea3nt;
M
0.3 Canids relative molar sizes are regulated by an inhibitory cascade,
Omnivorous indicating that there are single mechanisms that inhibit
0.2
distalmolars(Kavanaghet al.2007;Polly2007).Therefore,
0.1
66% variationinrelativemolarsizesincanidspeciesisalsoregu-
0.0
Faunivorous lated by an inhibitory cascade. In this study, carnivorous
0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0
species tended to have lower M /M scores (or relatively
M2/M1 2 1
larger M ) than omnivorous species. Moreover, the two
1
carnivorousspecieshavingthesmallestM inrelationtoM
2 1
80% have lost M . Carnivorous species exhibited the pattern
3
Carnivorous Omnivorous Insectivorous M >> M >> M , but omnivorous species exhibited
1 2 3
M > M > M .Asthenumberofinsectivorousspecieswas
Figure1. Variation in relative molar sizes among canid species. 1 2 3
Interspecific variation in canids, excluding Otocyon megalotis (red limited, they could not be analyzed statistically; however,
line), differed from that in murine rodents (blue line). Colors and thesespeciestendedtohavemoreequal-sizedmolars.Thus,
shapes indicate diet of a given species (red square: carnivorous, blue the inhibitory cascade reflects dietary adaptation in canid
circle:omnivorous,greentriangle:insectivorous)(Table1).(a) Species molars (Fig. 1). The relationships among inhibitory cas-
plots with standard deviations (SD). Numbers indicate species as in
cade, relative molar sizes, and dietary adaptation in canids
Table1.(b)DietarypatternswithSD.Occlusalviewofmolarrowsof
are similar to those in murine rodents (Kavanagh et al.
species for each diet in canids and murine rodents. Illustration of
2007). However, the patterns of adaptation differ between
murine rodents after Kavanagh etal. (2007). Maximum relative M
1
size reaches 80% of the total molar row in canids, and 66% in canids and murine rodents. For example, insectivorous
murinerodents. canids and herbivorous murines have equal-sized molars,
Table2. Regressionresults(RMA)oftheM /M versusM /M morphospace,showingconfidenceintervals(CI).
2 1 3 1
Typesofvariation Slope CImax CImin Intercept CImax CImin r P N
Inhibitorycascademodel 2.00 (cid:3)1.00
Canidae(with3molars) Interspecific 0.45 0.515 0.376 (cid:3)0.08 (cid:3)0.037 (cid:3)0.104 0.91 0.000 24
Canidae(withoutO.megalotis) Interspecific 0.48 0.537 0.412 (cid:3)0.09 (cid:3)0.057 (cid:3)0.119 0.93 0.000 26
Canidae(ondiet) Diet 0.48 0.538 0.438 (cid:3)0.09 (cid:3)0.061 (cid:3)0.111 0.99 0.035 3
Canislatrans Individual 0.52 0.632 0.380 (cid:3)0.10 (cid:3)0.041 (cid:3)0.138 0.41 0.003 51
Canislupus Individual 0.49 0.606 0.267 (cid:3)0.07 0.003 (cid:3)0.101 0.69 0.000 28
Canismesomelas Individual 0.71 2.059 0.240 (cid:3)0.16 0.011 (cid:3)0.649 0.43 0.060 20
Lycalopexgriseus Individual 0.46 0.594 0.184 (cid:3)0.09 0.036 (cid:3)0.163 0.60 0.018 15
Urocyoncinereoargenteus Individual 0.94 1.150 0.676 (cid:3)0.36 (cid:3)0.220 (cid:3)0.466 0.54 0.002 31
Vulpeslagopus(with3molars) Individual 0.73 0.946 0.425 (cid:3)0.16 (cid:3)0.052 (cid:3)0.233 0.78 0.001 15
Nyctereutesprocyonoides(with3molars) Individual(smallisland) 0.82 2.096 0.560 (cid:3)0.32 (cid:3)0.186 (cid:3)0.984 0.55 0.036 26
ª2012TheAuthor.PublishedbyBlackwellPublishingLtd. 281
UniqueInhibitoryCascadeinCanidMolars M.Asahara
(a) (b)
Normal Normal
* *
Missing Missing
0.15
0.15
0.10
M1 0.10
3/
M
0.05
0.05
0.00 0.00
0.25 0.30 0.35 0.40 0.40 0.45 0.50 0.55
M2/M1 M2/M1
Figure2. Individualvariationinrelativemolarsizesin(a)Vulpeslagopusand(b)Nyctereutesprocyonoides,indicatingpresenceorabsenceofM
3
by color (black: missing on both sides; dark gray: missing on one side; right gray: normal). Within species, M /M scores differed significantly
2 1
betweennormalindividualsandindividualsmissingM (indicatedbyasterisks).
3
and carnivorous canids and faunivorous murines (eating
Variability in relative molar sizes and loss of
animals including insects) have relatively larger M . This
1 M3 in canids
may be due to the difference in absolute body size
among the species. Insects are sufficiently large prey for Theresultsofindividualvariationclearlyshowedthecorre-
murines, and these mammals need to concentrate their lationsbetweenM /M andM /M (Table 2),anddifferent
2 1 3 1
masticatory function on one majortooth. Similarly, mam- M /M scores between individuals having M vs.individu-
2 1 3
malian flesh is sufficiently large to be accommodated by alsinwhichM wasmissing(Fig. 2),indicatingthatgreater
3
canid molars. In contrast, insects are small food items for inhibition results in smaller distal molars and/or a loss of
canids, and canids require a long molar row with equal- M . Although the correlation coefficient r was not high
3
sized teeth in order to chew a number of insects at once between individuals (Table 2), individual variation reflects
(Ungar2010). a large number of environmental factors. Therefore, a sig-
Among the species examined, only O. megalotis had nificant correlation indicates that individual variation
equal-sized molars, and was located distantly from the reflectsaninhibitorycascade.Individualvariationindicates
other canids in morphospace (Fig. 1). This may be related the variability within a species (Klingenberg 2005); there-
to the unique characteristic in O. megalotis of having four fore, canid species differ from murine rodents in terms of
lower molars (Sillero-Zubiri 2010; Ungar 2010). Interspe- variability, that is, they have lower slopes (Table 2). These
cific variation in the other canids exhibited a unique patterns of variability are likely to be the source of the
pattern of molar ratios that differs from that in murine uniqueinterspecificvariationobservedincanids.
rodents (Kavanagh et al. 2007), indicating a difference in The results of interspecific and individual variation in
the inhibitory cascade. Such differences have been M loss indicate that M loss in canids must be generated
3 3
reported in previous studies; however, the slope of the by greater inhibition during evolution. Individual plots
difference in canids was lower than that in any previously for V. lagopus provided a good illustration of interspecific
reported taxa (canids, interspecific: 0.48; canids, individ- variation; V. lagopus individuals with missing M were
3
ual: 0.46–0.94; murine and arvicoline rodents and South plotted near C. alpinus, a species that has lost M ,
3
American ungulates: 1.17–2.15) (Kavanagh et al. 2007; whereas individuals with normal dentition were plotted
Renvois(cid:1)e et al. 2009; Wilson et al. 2012). In mouse exper- near species that retain M (e.g. Vulpes vulpes, Vulpes
3
iments, all inhibition molecules were eliminated and macrotis) (Figs. 1 and 2). Therefore, V. lagopus may be in
interspecific variation in murine rodents was identical to a so-called ‘transitional stage’ of evolution of dental
observed variation in molar proportions (Kavanagh et al. formulae, reflecting greater inhibition of the inhibitory
2007). However, diffusion patterns may differ between cascade along the trajectory of carnivorous adaptation
inhibition molecules. It is possible that particular mole- (Fig. 3). Nevertheless, the N. procyonoides population
cules with low diffusion efficiency have high evolvability exhibited a high frequency of missing M despite
3
and generate unique slopes in canids; however this is not relatively lower inhibition in relation to V. lagopus or
yet clear. other canids that retain M (Figs. 1 and 2). This
3
282 ª2012TheAuthor.PublishedbyBlackwellPublishingLtd.
M.Asahara UniqueInhibitoryCascadeinCanidMolars
22: Vulpes lagopus
23: Vulpes macrotis
26: Vulpes vulpes
27: Vulpes zerda
16: Nyctereutes procyonoides
17: Otocyon megalotis
19: Urocyon cinereoargenteus
1: Atelocynus microtis
7: Cerdocyon thous
11: Lycalopex griseus
12: Lycalopex gymnocercus
13: Lycalopex sechurae
8: Chrysocyon bracchyrus
18: Speothos venaticus
2: Canis adustus
3: Canis aureus
5: Canis lupus
4: Canis latrans
9: Cuon alpinus
15: Lycaon pictus
0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0
M2/M1
Figure3. M /M scoresforeachspecies,andtheirphylogeneticrelationshipsfromBardelebenetal.(2005).Colorsandfiguresindicatedietasin
2 1
Fig.1.NumbersindicatespeciesasinTable1.
N. procyonoides population has probably been affected by relation to other canid molars, or all molars of rodents;
inbreeding depression, as the population examined was therefore, if analysis is based on the three-dimensional
introduced from the mainland to a small island (Saeki tooth volume, the change in the canid M must become
1
2009). The molar proportion of this population over- further amplified.
lapped with that of the mainland populations that retain Canids have evolved different proportions among func-
M (personal observation). Geographical isolation and tionally distinct parts of their lower molars, that is, the
3
fixation of series of genes that relate to the inhibitory cas- shearing surface (trigonid of M ), which is important for
1
cade or other mechanisms could also be an important a carnivorous diet, and the grinding surface (talonid of
process in the evolution of dental formulae (Asahara M , M , and M ), which is important for omnivorous
1 2 3
et al. 2012). The fact that no V. lagopus and N. procyono- and insectivorous diets (Van Valkenburgh and Koepfli
ides individuals exhibited M /M scores of lower than 1993; Friscia et al. 2007). Therefore, the particular pattern
3 1
0.03 and 0.05, respectively, may relate to additional of inhibitory cascade (with amplified change in M )
1
mechanisms for the regulation of M development and a would contribute to dramatic changes in the proportion
3
possible threshold for M development or loss; teeth of shearing surface in M versus grinding surface in M
3 1 2
germ which are smaller than some threshold at the criti- and M (i.e. dramatic changes in function) over the
3
cal stages cannot continue to develop into mature teeth course of evolution. This dramatic change is regulated by
(Gruneberg 1951; Wolsan 1989; Szuma 2003). a single developmental mechanism, the inhibitory cascade.
Therefore, the molars of canids can readily evolve to
adapt to a carnivorous, omnivorous, or insectivorous diet,
Functional consequences of the unique
and canids thus have the potential for evolutionary plas-
inhibitory cascade pattern in canids
ticity in their diet. These patterns of variation must have
contributing to the evolvability of diet
contributed to the diversity of dietary patterns and their
The patterns of interspecific variation shown here are parallel evolution among canids (Fig. 3) (Goswami 2010;
indicative of a unique inhibitory cascade pattern with less Sillero-Zubiri 2010), and to the short-time divergence
steep regression lines in the morphospace (Fig. 1, and diversity of dietary patterns in Lycalopex species (Pe-
Table 2) than any other previously reported mammals rini et al. 2009). Polly (2007) inferred the existence of
(Kavanagh et al. 2007; Renvois(cid:1)e et al. 2009; Wilson et al. inhibitory cascade regulation across all mammals. In addi-
2012). Guided by this pattern, the change in relative size tion, previous studies (Renvois(cid:1)e et al. 2009; Wilson et al.
of M has been amplified in canids (e.g. M comprises 2012), and my results, suggest that patterns of the inhibi-
1 1
>80% of the total molar surface in S. venaticus, whereas tory cascade can differ among taxa. Moreover, I suggest
the maximum proportion occupied by M in murine that these different inhibitory cascade patterns have con-
1
rodents is 66%; Fig. 1). My analysis is based on the two- tributed to different evolvability and diversity of diet
dimensional occlusal surface, as used in previous studies among taxa. That is, clade-specificmodificationindevelop-
(Kavanagh et al. 2007; Renvois(cid:1)e et al. 2009; Wilson et al. mental mechanisms could have promoted the capacity for
2012). However, the canid M is a high cuspid tooth in dietary adaptation; that is, the dynamics of proximate and
1
ª2012TheAuthor.PublishedbyBlackwellPublishingLtd. 283
UniqueInhibitoryCascadeinCanidMolars M.Asahara
ultimatefactors.Investigationofinhibitorycascadepatterns Hammer,D.A.T.Harper,andP.D.Ryan.2001.PAST:
in other mammals will further our understanding of these paleontologicalstatisticssoftwarepackageforeducationand
evolutionarydynamics. dataanalysis.Palaeontol.Electronica4:9.
Harjunmaa,E.,A.Kallonen,M.Voutilainen,K.H€am€al€ainen,
M.L.Mikkola,andJ.Jernvall.2012.Onthedifficultyof
Acknowledgments
increasingdentalcomplexity.Nature483:324–327.
IthankEileenWestwig,TeresaPacheco,andDarrinLunde Holliday, J. A. 2010. Evolution in Carnivora: identifying a
to arrange specimens in AMNH. I thank Laura Wilson for morphological bias. Pp. 189–224 in A. Goswami and A.
helpfulproofreadingandcommentsonearlyversionofthe Friscia, eds. Carnivoran evolution: new views of phylogeny,
manuscript,andMasaharuMotokawaandmembersofthe form and function. Cambridge University Press,
LaboratoryofSystematicZoologyinKyotoUniversity,and Cambridge.
Yoshihiro Morishita and members of the Laboratory for Kavanagh,K.D.,A.R.Evans,andJ.Jernvall.2007.Predicting
Developmental Morphogeometry, RIKEN CDB for helpful evolutionarypatternsofmammalianteethfrom
development.Nature449:427–432.
comments. I thank anonymous editors and reviewers for
Kirschner,M.,andJ.Gerhart.1998.Evolvability.Proc.Natl
helpful suggestions to improve the manuscript. This study
Acad.Sci.95:8420–8427.
was supported in part by the Mishima Kaiun Memorial
Klingenberg,C.P.2005.Developmentalconstraints,modules,
Foundation, The Kyoto University Foundation, the Global
andevolvability. Pp.219–248inB.HalligrimssonandB.K.
COE Program A06 (MEXT, Japan), JSPS AA Science Plat-
Hall,eds.Variationacentralconceptofbiology.Elsevier
formProgram,JSPSResearchFellowshipsforYoungScien-
AcademicPress,Burlington,USA.
tists,andGrant-in-AidforJSPSFellows(11J01149).
Laffont,R.,E.Renvois(cid:1)e,N.Navarro,P.Alibert,andS.
Montuire.2009.Morphologicalmodularityandassessment
Conflict of Interest ofdevelopmentalprocesseswithinthevoledentalrow
(Mictorusavails,Arvicoline,Rodentia).Evol.Dev.
None declared.
11:302–311.
Laland,K.N.,K.Sterelny,J.Odling-Smee,W.Hoppitt,and
References
T.Uller.2011.Causeandeffectinbiologyrevisited:is
Asahara,M.,A.Kryukov, andM.Motokawa.2012.Dental Mayr’sproximate-ultimatedichotomystilluseful?Science
anomaliesintheJapanesemoleMogerawogurafrom 334:1512–1516.
northeastChinaandthePrimorskyregionofRussia.Acta Ohtaishi,N.1986.Honyuruinoha.Pp.123–134inM.Goto
Theriol.57:41–48. andN.Ohtaishi,eds.Comparativeodontology,morphology,
Bardeleben,C.,R.L.Moore,andR.K.Wayne.2005.A functionandevolutionoftoothinvertebrates. Ishikawa
molecularphylogenyofthecanidaebasedonsixnuclear publishers,Inc.,Tokyo(inJapanese).
loci.Mol.Phyl.Evol.37:815–831. Perini,F.A.,C.A.Russo,andC.G.Schrago.2009.The
Barton,N.H.,D.E.Briggs,J.A.Eisen,D.B.Goldstein,and evolutionofSouthAmericanendemiccanids:ahistoryof
N.H.Patal.2007.Evolution.ColdSpringHarbor rapiddiversificationandmorphologicalparallelism.J.Evol.
LaboratoryPress,NewYork,USA. Biol.23:311–322.
Benton,M.J.2004.Vertebratepalaeontology, 3rded. Polly,P.D.2007.Development withabite.Nature
BlackwellPublishing,Malden. 449:413–415.
Friscia,A.R.,B.VanValkenburgh,andA.R.Biknevicius. Popowics,T.E.2003.Postcaninedentalforminthe
2007.Anecomorphologicalanalysisofextantsmall mustelidaeandviverridae(Carnivora:Mammalia).
Carnivorans.J.Zool.272:82–100. J.Morph.256:322–341.
Futuyma,D.J.2010.Evolutionaryconstraintandecological Renvois(cid:1)e,E.,A.R.Evans,A.Jebrane,C.Labru(cid:3)ere,R.Laffont,
consequences.Evolution64:1865–1884. andS.Montuire.2009.Evolutionofmammaltooth
Giannini,N.P.,andN.B.Simmons.2007.Elementhomology patterns:newinsightsfromadevelopmentalprediction
andtheevolutionofdentalformulaeinmegachiropteran model.Evolution63:1327–1340.
bats(Mammalia:Chiroptera:Pteropodidae).Am.Mus.Nov. Saeki,M.2009.Nyctereutesprocyonoides(Gray,1834).Pp.
3559:1–27. 216–217inS.D.Ohdachi,Y.Ishibashi,M.A.Iwasa,
Goswami,A.2010.IntroductiontoCarnivora.Pp.1–24inA. T.Saitoheds.ThewildmammalsofJapan.SHOUKADOH
GoswamiandA.Friscia,eds.Carnivoranevolution:new BookSellers,Kyoto.
viewsofphylogeny,formandfunction.Cambridge Sillero-Zubiri,C.2010.FamilyCanidae(dogs).Pp.352–447
UniversityPress,Cambridge,UK. inD.E.WilsonandR.S.Mittermeier, eds.Handbookof
Gruneberg,H.1951.Thegeneticsofatoothdefectinthe themammalsoftheworldVol.1.Carnivores.Lynx
mouse.Proc.R.Soc.B138:437–451. Edicions,Barcelona.
284 ª2012TheAuthor.PublishedbyBlackwellPublishingLtd.
M.Asahara UniqueInhibitoryCascadeinCanidMolars
Szuma,E.2003.Microevolutionarytrendsinthedentitionofthe Wilson,L.A.B.2011.Thecontributionofdevelopmental
redfox(Vulpesvulpes).J.Zool.Syst.Evol.Res.41:47–56. palaeontologytoextensionsofevolutionarytheory.Acta
Ungar,P.S.2010.Mammalteeth:origin,evolution,and Zool.Inpress
diversity.TheJohnsHopkinsUniversityPress,Baltimore. Wilson,L.A.B.,R.H.Madden,R.F.Kay,andM.R.
VanValkenburgh,B.2007.De′ja‘vu:theevolutionoffeeding S(cid:1)anchez-Villagra.2012.Testingadevelopmentalmodelin
morphologiesintheCarnivora.Integr.Comp.Biol. thefossilrecord:molarproportionsinSouthAmerican
47:147–163. ungulates.Paleobiology38:308–321.
VanValkenburgh,B.,andK.P.Koepfli.1993.Cranialand Wolsan,M.1989.DentalpolymorphismintheGenusMartes
dentaladaptationstopredationincanids.Symp.Zool.Soc. (Carnivora:Mustelidae)anditsevolutionarysignificance.
Lond.65:15–37. ActaTheriol.40:545–593.
Wagner,G.P.,andL.Altenberg.1996.Complexadaptations
andtheevolutionofevolvability.Evolution50:967–976.
ª2012TheAuthor.PublishedbyBlackwellPublishingLtd. 285