Application of Genomic Analyses
on Local Animal
Genetic Resources
Martino Cassandro
Italy
Outline
Introduction
Genomic Assisted CONSERVATION Scheme for Local AnGR
Genomic Assisted CHARACTERIZATION of local AnGR
Genomic Assisted TRACEABILITY for local Animal Products
2
Introduction
Genetic Erosion of AnGR:
• 90% of animal food derived by only 14 species (FAO, 1999)
• 16% of animal domestic breeds are disappearances in the last century
(Hall and Ruane, 1993).
• 30% of world breeds are in risk of extinction (Hammond, 1996) and
annual rate of extinction is increasing (FAO, 1995)
• Hammond and Leitch (1996) showed that the genetic variance among
breeds account to 30–50% of the total variance
3
Introduction
In Europe:
• almost 2/5 present breeds are on risk of extinction
• 1/3 of breeds are extincted from the beginig of XX century.
(Hammond and Leitch, 1996)
4
Introduction
•
Why Animal Biodiversity is so important and large in Italy:
1. Area refuge after last glaciation
2. Large variety of environments (Alps, Hills, Plains, Coasts, Islands)
3. A rich history of migrations of human and animals populations
4. Animal biodiversity as tool and to valorize animals products, local/niche
markets, preserve territory and culture, and population history
5. Over 57,000 animal species (and 8,000 plant species) are the numbers
that make the Italy as the European nation's richest biodiversity.
5
Introduction
•
Why Genomic Approach is useful for “Local” Animal Biodiversity
1. No pedigree information are available
2. Absence of studies on performances/characterization
3. No reliable information within and among breeds/populations
4. Scarce economic interest to preserve original “genome”
6
Introduction
An important strategy
to INCREASE Add Value for Animal Products
to PRESERVE Environment and Biodiversity
to ORIENTATE Tourism & Food Consumptions
might be to “PROMOTE” the LINK
LOCAL BREED
TERRITORY
PRODUCT
7
Farmer
Animal
Traditional Yield Chain
Milk-Meat
Processed Products
Commercial and Marketing
Modern Integrated Chain
Grassland-Pastures
Environment - Landscape
Tourism/quality of life
Local Economy
8
Farmer
Animal
Modern Integrated Chain
Traditional Yield Chain
Environmental Chain
Grassland-Pastures
Milk-Meat
Processed Products
Commercial and Marketing
Environment - Landscape
Tourism/quality of life
Local Economy
9
Farmer
Animal
Modern Integrated Chain
Environmental Chain
Traditional Product Chain
Grassland-Pastures
Milk-Meat
Processed Products
Commercial and Marketing
Local breeds – Landscape
Tourism/quality of life
Local Economy
10
11
Introduction
Chianina-Romagnola
Consortium 5R
Marchigiana-Podolica Italian Meat
Maremmana
Reggiana
Parmigiano
Reggiano
di razza Reggiana (37 euro/kg)
Burlina
Veneto Sheep
Breeds
Morlacco
Pitina
12
Introduction
BREED-PRODUCT LINK
Cinta Senese
Valdostana
Lardo di Colonnata
Fontina
Rendena
Padovana Chicken
breed
Cheese
Rendena
Pro Avibus
Nostris
13
Conservation of Local
Poultry Breeds
Conservation of Local Poultry Breeds
Robusta Maculata (RM)
Robusta Lionata
(RL)
Pépoi (PP)
Padovana Dorata (PD)
Ermellinata di Rovigo (ER)
Common chicken
Padovana chicken
15
Genomic Conservation of AnGR
•
COVA Project, since 2000. 11 Chicken local breeds (+ other 3 avian species)
•
An In situ conservation scheme based on 3 breeding units/breed
•
Each nucleus breed, within breeding unit, is based on 20 M + 34 F
•
Males are divided in 2 families
•
Females of the same breed, within breeding unit, are reared all together
•
Individuals are genotyped extracting genomic DNA by whole blood
•
Individuals are selected in order to:
– phenotypic standard breeding
– family origin
– individual contribution at overall Vg
16
MARKER ASSISTED CONSERVATION SCHEME
BIANNUAL CHANGE OF ALL ANIMALS
10 old males
17 old females
20 males
34 females
210 eggs by first family – period 1
210 eggs by second family –period 2
420 eggs in 6 weeks
0 months
0,5 months
4 months
50% N/I
210 chicks born alive
105 males
5 males/family
10 young males
5 months
105 females
6/7 months
8/9 females/family
17 young females
8/12 months
Annual Males Rotation Scheme
BREEDING UNIT (BU) n. 1 (TV)
Males
Selected
BU N.2 (BL)
1
0,9
BU N. 3 (RO)
BU, n. 3
0,8
TREND OVER YEAR OF QUOTA OF MALE’S ORIGIN
0,7
0,6
0,5
33%
0,4
0,3
0,2
BU, n. 2
0,1
BU, n. 1
0
1
2
3
4
YEAR
5
6
7
18
Genomic Characterization of AnGR
Breed
Code
Alleles/
breed
HE (nb)
±SD
HO
±SD
FIS
fij breed
Brown Layer
BL
3.80
0.559 0.141
0.622 0.233
-0.117***
0.439
Ermellinata di Rovigo
ER
3.14
0.420±0.175
0.384±0.248
0.089***
0.573
Pèpoi
PP
2.51
0.243±0.239
0.240±0.236
0.018*
0.769
Robusta Lionata
RL
2.43
0.367±0.229
0.317±0.264
0.131***
0.657
Robusta Maculata
RM
2.17
0.293±0.225
0.292±0.226
0.003
0.721
Polverara Bianca
PB
3.01
0.436±0.190
0.366±0.201
0.161***
0.577
Polverara Nera
PN
3.45
0.463±0.177
0.413±0.170
0.109***
0.559
Padovana Camosciata
PC
2.27
0.305±0.257
0.287±0.271
0.062
0.704
Padovana Dorata
PD
2.66
0.340±0.199
0.329±0.230
0.034
0.689
Breed
PB (36)
PN (52)
RL (43)
RM (45)
ER (45)
PD (24)
PC (26)
PP (45)
19
Genomic Characterization of AnGR
Structure analysis of six Italian local chicken breeds assuming K = 6, 7, 8, 9, 10.
Only most probable solutions for each K are shown.
20
Genomic Characterization of AnGR
N-J tree drawn by DR distances (1000 bootstrap repetitions)
21
Genomic Characterization of AnGR
22
N-J tree drawn by Dk distances among nuclei
Genomic Characterization of AnGR
23
N-J tree drawn by Dk distances among nuclei
Genomic Characterization of AnGR
Alpagota
Fiemmese
Foza
Bergamasca
Brogna
Suffolk
24
Genomic Characterization of AnGR
Breed
ALPAGOTA
BERGAMASCA
BROGNA
FIEMMESE
FOZA
LAMON
SUFFOLK
n.
individuals
75
30
50
30
35
21
33
Average
TOTAL
274
Het
Ho
0,673 ± 0,122
0,783 ± 0,082
0,734 ± 0,097
0,758 ± 0,115
0,677 ± 0,208
0,760 ± 0,067
0,701 ± 0,142
0,719 ± 0,137
0,750 ± 0,107
0,746 ± 0,113
0,781 ± 0,121
0,726 ± 0,231
0,818 ± 0,076
0,673 ± 0,143
N. average
alleles
6,7 ± 2,3
8,1 ± 1,7
7,1 ± 1,7
8,3 ± 2,8
6,5 ± 1,6
6,3 ± 1,8
7,1 ± 1,5
0,727 ± 0,043 0,745 ± 0,046
7,2 ± 0,77
0,783 ± 0,087 0,737 ± 0,096
12,2 ± 3,2
Allelic
richness
5,4 ± 1,4
7,6 ± 1,4
6,3 ± 1,6
7,6 ± 2,6
6,2 ± 1,6
6,3 ± 1,8
6,5 ± 1,3
6,6 ± 0,8
8,3 ± 1,9
25
Genomic Characterization of AnGR
■Alpagota, ▲Bergamasca, ▲Brogna, ● Fiemmese, ■ Foza, ■ Lamon, ● Suffolk
0 .2
ALP1
ALP31
Neighbour-Joining Dendrogram using “pairwise” distances
26
Genomic Characterization of AnGR
AALLP
P22
ALP377
FIE19
FIE
ALP23
ALP 12
ALP 36
35
AALLPP26
ALP 206
AL 204
AL P205
AL P214
P2
17
AALLP
AL P2405
3
P
AL 20
AL P2 8
A P2 43
ALLP2 01
A P 22
BRLP 209
LA O2225
A M1 34
A LP 41
ALP23
ALP1
ALP31
ALP28
ALP2
P412
AL
6
ALP220
■Alpagota, ▲Bergamasca, ▲Brogna, ● Fiemmese, ■ Foza, ■ Lamon, ● Suffolk
0 .2
A L 1
A LP P2 1
AL LP 22 2
2 3
A P2 24
LP 0
39 7
4
R3
BE R24
BE R16
BE R35
BE R30
BE R26
BE ER62
B R2
BE R57
BER1 9
BE ER12
B R 20
BEER 25
B ER 107
B R3 3
BEO2 23 2
BR RO O8
B R
B
UPGMA dendrogram using “pairwise” distances
0
P4 13
AL LP P17 8
A L P1 8
A L 3 1
A LP 21
A LP 19
A LP 41
A LP 53
A LP 210
A LP 21
A LP2 25
A LP 02
A LP2
A LP5
8
A
BRBR
B O2O7
BRBR RO8 41 2
O 1
O
B BR 2 71
BRRO1O5 44
O2 42 9
B
B R 4
BRRO2O74 2
B O2340
BRRO7 5
O 6
BRBRO 69
BR O2362
O
9
BR 238
BR O78
BROO68
BRO 67
P
AL P55
AL P29
AL P30
AL P32
AL P33
AL P9
AL 21
ALP 20
ALP 6
ALP 6
ALP116
ALP2 4
ALP3
ALP51
ALP215
ALP52
ALP10
ALP42
ALP218
BRO236
ALP47
ALP54
ALP56
ALP4
ALP24
ALP213
BR 73
BR O58
BR O94
BRO O93
2
BRO 62
BRO 79
BRO895
BER335
SUF26
SUF30
SUF25
SUF4
SUF18
SUF32
SUF29
SUF7
SUF31
SUF23
SUF28
5
SUF1
1
SUF1
SUF94
SUF1F3
SU F1
SU F8
SU 17
SUFF10
S U F2 7
SU F13
SU F162
SU O9
BR
111
ER
Z10 0
FBO Z26
FO Z9163
O 1
FFOZZ1101
FO 25
FOZZ250
FO Z107
FO Z254
FO 11
FOZZ1115
FO Z109
FO 120
FOZ
FOZ143
FOZ108
FOZ118
FOZ259
FOZ993
FOZ10
FOZ261
FOZ105
FOZ253
FOZ
986
FO
Z10
FOZ257
FOZ256
FOZ117
FOZ10
FOZ2 4
FOZ1 52
FOZ 00
FOZ1258
19
27
Genomic Conservation of AnGR
28
Genomic Conservation of AnGR
A
B
Family #1
C
D
Family #2
29
Local Cattle Breed - BURLINA
“Morlacco cheese of Burlina cattle breed”
31
Economic Characterization of AnGR
400
“Economic comparison Burlina vs HF”
Diff. in profit, €/year per cow
200
0
€0
-200
-400
-€ 321
-600
-800
-€ 566
-€ 766
-1.000
Regional Premium
no
200 €/capo
200 €/capo
200 €/capo
Milk valorization
no
no
+ 0,05 €/kg
+ 0,05 €/kg
9079 kg
9079 kg
9079 kg
7706 kg
Holstein Friesian yield
32
Genomic Characterization of AnGR
Assignment of individuals at 4 clusters breeds using software STRUCTURE 2.1,
Group #1 : unknown BREED?
Group #2: Burlina breed
Group #3: Bruna breed
Group #4: Frisona breed.
Unknown
Breed ?
BURLINA
BROWN
SWISS
HOLSTEIN
FRIESIAN
33
Outline
Introduction
Genomic Assisted CONSERVATION Scheme for Local AnGR
Genomic Assisted CHARACTERIZATION of local AnGR
Genomic Assisted TRACEABILITY for local Animal Products
34
Genomic Traceability of AnGR
TRACEABILITY is an important tool in EU not only
for Consumers but also for Producers
No. of Certified Products
PDO (DOP) e PGI (IGP)
1° ITALY
2° France
3° Portugal
4° Spain e Greece
6° Germany
145
141
92
84
67
Loss of 2,5 billion €/year
for FOOD PIRACY*
Example of False ITA-Products :
- Asiago by Wisconsin (USA)
- Parma Ham (USA)
- Parmesao (Brasile)
- Regianito (Argentina)
* dati CIA e Legambiente
35
Genomic Traceability of AnGR
The ability to trace the history and to trace the
application or location might be divided in
Traceability back to the
origin
Production systems
including feeding diets
Geographical origin
Individual
Breed
Species
Traceability
of process
Genetic
Processing
Conservation processes
Adulteration of products
36
Genomic Traceability of AnGR
Traceability not only in EU
Mandatory only in the E.U. and Japan
Mandatory (at least for exported beef) in Brazil, Australia,
Argentina and Canada
Voluntary in the U.S.A.
37
Genomic Traceability of AnGR
DNA is:
 Inalterable
 Present in every cell
 Persistent to various treatment
Widely used of molecular markers:
 SNP (single nucleotide polymorphism)
 SSR (eg. microsatellites)
 Whole Genome
38
Genomic Traceability of AnGR
Individual Genetic Fingerprint is constant
39
Genomic Traceability of AnGR
Probabilistc approach
based on allelic frequencies
Match Probability inferior to 1*E-6
(Weir, 1996)
m = number of loci
n = number of alleles at locus k
pki(j) = Allelic frequency of the allele i (j) at locus k
40
Genomic Traceability of AnGR
Collection of milk and
blood from:
Collection of muscles
and blood from:
Piemontese n=24
Holstein
Friesian n=41
Chianina n=24
Brown Swiss
n=53
Marchigiana=22
41
Genomic Traceability of AnGR
N° loci
12
Dairy
breeds
1.57E-10
Beef
breeds
1.89E-11
All breeds
1.13E-11
8
7.02E-09
2.23E-09
1.27E-09
4
1.25E-05
1.26E-05
6.76E-06
1
5.3E-02
3.64E-02
2.74E-02
42
Genomic Traceability of AnGR
Breed traceability approaches
Deterministic approach:
Probabilistic approach:
allelic variants fixed in different
breeds (MC1R,myostatin)
allelic frequencies typical in
different breeds
High cost of analyses
BUT
Product-breed link can improve breed economic profitability
43
Breed Traceability in Cattle (15 microsatellite)
Probabilistic Approach
Breed assignement
CHI
CHI
98 %
ROM
2%
MAR
ROM
MAR
PIE
HF
2%
-
-
90 %
8%
-
8%
88 %
2%
2%
PIE
-
2%
5%
93 %
-
HF
-
-
-
-
100%
(Ciampolini e coll., 1999)
44
Breed Traceability in Sheep (10 microsatellite)
% of Breed Assigment – Probabilistic Approach
%
ALPagota
N.
ALP BER BRO FIE FOZ SUF 252
1
3
3
93
75
83
BERgamasca
BROgna
6
FIEmmese
3
FOZa
3
SUFfolk
3
3
7
92
2
94
3
7
3
91
3
97
30
50
30
35
32
45
Breed Traceability for Cheese
Deterministic Approach
using MC1R e KIT genes
Brown Holstein Simmental Reggiana
(Russo e Fontanesi, 2004)
46
Model of inheritance of the
Extension (E) locus and Spotted (S) locus
MCR1 gene
KIT gene
(Melanocortin receptor 1)
Alleles
Alleles
ED
E+
e
S = not spotted
s = spotted
= Black
= Brown
= Red
Genotypes
ED ED Black Holstein
ED E+ Black
ED e Black
E+ E+ Brown Brown Swiss
E+ e Brown
e e
Red
Simmental
Reggiana
Genotypes
SS not spotted Brown Swiss,
Reggiana
Ss not spotted
ss spotted
Holstein
Simmental
47
Breed Traceability for Cheese
Distinction of the cheese yielded with milk of different breeds by
Locus Extension (E)
and
Locus Spotted (S)
MCR1 gene
KIT gene
(Russo e Fontanesi, 2004)
48
Breed Traceability in Pigs
AFLP (E12/T14) of 3 PIGS (2 DUROC and 1 IBERIAN)
Duroc
Duroc
Iberian
49
Relationship between the additional cost of DNA traceability
and the type of existing traceability system
50
Genomic Traceability of AnGR
Genome wide information is necessary for several aims for AnGR:
- Conservation
- Characterization
- Traceability
- Authenticity
…....and other traditional/already known uses as
- Paternity check
- Re-construction of pedigree
- Heterosis prediction
- Other uses in animal breeding
51
Conclusioni
• Altri utilizzi delle informazioni molecolari in
animal breeding:
–
–
–
–
Tracciabilità
Controllo paternità
Ricostruzione di pedigree
Previsione dell’eterosi
• Altri utilizzi delle tecniche statistiche utilizzate in
animal breeding
–
52
Thank for the attention
A FUTURE PERSPECTIVE IS A
MULTIFUNCTIONAL USE OF GENOME WIDE INFORMATION
53
Outline
Introduction
Genomic Characterization of AnGR in the Veneto Region (ITA)
• Bovine local breed (BURLINA)
• Sheep breeds (ALPAGOTA, FOZA, LAMON e BROGNA)
• Avian species (CHICKEN, DUCK, TURKEY and GUINEA FOWL)
Genomic Conservation of AnGR in the Veneto Region (ITA)
• Burlina identification
• Alpine sheep breeds
• Co.Va. Project
Genomic Traceability of AnGR in the Veneto Region (ITA)
• Species/Breed/Individual level
• Meat/Milk/Cheese products
54
Genomic Characterization of AnGR
INDIVIDUAL IDENTIFICATION/SELECTION
at the birth
at about 5/6 months of age
Using a threshold index based on:
• family origin
• breed’s standards
• production performance
• reproduction performance
Cluster analyses for individual selection
within family
Family 1
Family 2
Win tag
55
CONSERVAZIONE
Progetto Co.Va - 2
Le indicazioni dei marcatori molecolari (DNA microsatellite) vengono usate
per determinare le distanze genetiche e per monitorare il trend di variabilità
genetica nelle popolazioni
Per mantenere la variabilità genetica è stato importante usare per lo schema
di riproduzione i maschi che, intra gruppo, hanno mostrato le più grandi
distanze genetiche.
Secondariamente, i polli selezionati devono rispettare gli standard di razza.
Infine vengono considerate le performance produttive e riproduttive.
Questi indici soglia permettono l’identificazione degli animali migliori per
sostituire la generazione precedente.
56
DIVERSITÀ E VARIABILITÀ GENETICHE
La variabilità tra le popolazioni (specie, razze) può essere valutata tramite
strumenti matematici, che traducono le differenze ad una misura di
distanza tra una coppia di popolazioni (Eding & Laval, 1999).
La diversità genetica è definita come “somma di informazione genetica
contenuta nei geni” (Pearce & Moran, 1994).
Le distanze genetiche più utilizzate fanno uso di differenze nelle
frequenze alleliche nelle diverse popolazioni. In linea di massima si può
utilizzare qualunque gene che mostri polimorfismi (Eding & Laval, 1999).
57
DIVERSITÀ E VARIABILITÀ GENETICHE
Misure di diversità genetica - 2
FIS (coefficiente di inbreeding) descrive la varianza genetica tra
individui in relazione alla (sotto)popolazione di appartenenza, ovvero
l’eccesso di omozigoti all’interno delle sottopopolazioni.
FIT: esprime il generale eccesso di omozigoti nella popolazione totale.
GST (coefficiente di differenziazione genetica) (Nei 1973) è analogo al
FST per un numero finito di sottopopolazioni, ma si basa solo sulle
frequenze alleliche senza che siano noti i genotipi individuali (Balloux
& Moulin, 2002).
Esistono analoghi che forniscono stime indipendenti dalla numerosità
del campione (Nei, 1987).
58
DIVERSITÀ E VARIABILITÀ GENETICHE
Distanze genetiche – 1
Esistono numerosi tipi di distanze genetiche, la cui interpretazione
dipende sempre dal modello di divergenza ipotizzato:
D (distanza genetica standard di Nei) (1972) è basata sul modello
di mutazione IAM.
Assume la contemporanea azione della mutazione (stesso tasso in
tutti i loci) e della deriva genetica corrispondenti a tempi di divergenza
molto grandi.
Il valore di D ha un andamento lineare nel tempo, tuttavia la linearità
viene meno nel caso dei microsatelliti a causa del loro alto tasso di
mutazione.
59
DIVERSITÀ E VARIABILITÀ GENETICHE
Distanze genetiche - 2
Goldstein (1995) e Shivers (1995) idearono nuove misure di diversità
genetica ((δµ)2)che si adattassero all’alto tasso di mutazione caratteristico
dei microsatelliti.
Takezaki e Nei (1996) hanno dimostrato che la distanza DC (CavalliSforza, 1987) e DA (Nei, 1987), sebbene non rispettino la linearità in
funzione del tempo, danno risultati migliori in termini di topologia degli
alberi filogenetici da esse derivati, probabilmente a causa della minore
varianza ad esse associata (Ending & Laval, 1999).
DC è basata sull’unica assunzione che la popolazione sia sottoposta
all’influenza della deriva genetica; le popolazioni sono concettualizzate
come punti in uno spazio multidimensionale e la distanza ricavata è di tipo
geometrico.
60
DIVERSITÀ E VARIABILITÀ GENETICHE
Distanze genetiche - 3
Nel caso di tempi di divergenza piuttosto recenti (differenziazione
al livello di razza), si può ammettere che l’effetto mutazione sia
trascurabile e che il fattore principale che descrive la variabilità
genetica sia la deriva genetica casuale.
La distanza di Reynolds DR soddisfa quest’ipotesi, sotto il modello di
mutazione IAM.
61