cacao flavor through genetics—anatomy of fine flavor · 2014. 12. 2. · 26 november 2014 † the...

The Manufacturing Confectioner • November 2014 25

Within the cocoa world, there is agreat deal of concern about boththe cocoa supply and the sustainability ofcocoa farmers the world over. Initiativessuch as CocoaAction within the WorldCocoa Foundation, as well as large cocoaand chocolate company programs, focuson long-term solutions to sector problems.One of the critical elements is the need toextend resistance to the major diseases ofcacao to all of the growing regions. At thesame time, these initiatives are not able tofocus on maintaining the suite of flavorsthat are part of the heritage of cacao as weknow it today. Flavor, with all of its diver-sity, is extremely hard to include within abreeding program for cacao. It is for this

reason, and the fear that not including fla-vor as a breeding criteria will cause theloss of previous chocolate flavor diversity,that the Fine Chocolate Industry Associa-tion (FCIA) has launched the HeirloomCocoa Preservation Initiative.

We know that flavor in the finishedchocolate comes from a multitude ofsources: the genetics of the trees; terroir,including the land and soil conditions aswell as climate during pod maturation (5 1/2months long); fermentation and drying; thelocal microbiome where the fermentation isbeing carried out; and the factory processessuch as roasting, conching, etc.

This article will look at the genetics ofthe tree itself. Some of what we’re going totalk about is pretty obvious, and somenobody knows anything about, includingus. There is much to learn. The issue thatwe face is answering the question, Wheredo these unique and special flavors come

from? Frankly, we just don’t know rightnow. This goes back to the question thatbiologists have faced for years on mostcrops and keep debating, Is it nature or is itnurture? In our case, nurture would be thekind of things that the farmers do to thecocoa on the farm — the postharvest prac-tices as well as the climate and micro-biome. Ultimately, the question of where

Cacao Flavor Through Genetics —Anatomy of Fine FlavorThe goal of the Heirloom Cacao Preservation Initiative is topreserve the diversity of flavor for generations to come.Ed Seguine, Seguine Cacao; Lyndel Meinhardt, USDA-ARS

Ed Seguine is presi-dent of SeguineCocoa, which hefounded in 2013. Hehas more than 30years’ experience inthe chocolate industry.

Lyndel Meinhardt is aResearch Plant Pathol-ogist and ResearchLeader for the Sustain-able Perennial CropsLaboratory at USDAAgricultural ResearchService.

Fine Chocolate Industry Association (FCIA)is focused on supporting fine chocolateprofessionals, with the purpose of being an“unbiased collective voice of quality andinnovation that promotes fine chocolatemaking practices from blossom to bonbonand bar.” This couples farmer and environ-mental sustainability to the reason we are allhere: the beautiful diversity of flavor pre-sented by Theobroma cacao. Those whoshare a passion for the flavors of chocolate,farmer and environmental sustainability areencouraged to find information on joiningthe FCIA as a member or friend at their web-site. www.finechocolateindustry.org

26 November 2014 • The Manufacturing Confectioner

do these flavors come from — and how can we preservethem in the face of critically needed breeding for dis-ease resistance and productivity — is of paramountimportance to us, in particular, as fine chocolate mak-ers and to consumers everywhere who enjoy and loveall of the diversity of flavors in cacao.

While there are a multitude of factors that influencethe flavor, within the FCIA Heirloom Cacao Preserva-tion (HCP) program we standardize as much as possi-ble so that we’re treating these beans in an equivalentway in both recipes and processing. We do adjust theroasting process for the type of bean that we have, but,other than that, we are trying to standardize our pro-cessing so that we can then address what flavors arecoming inherently from the beans.

The HCP is important in preserving the diversity ofcocoa flavor. For the first time, we can understand andidentify tree varieties that are really unique in flavor. Itallows us to take information to the next level wherewe can analyze it to tease out what contributes to fla-vor. This is all being done at the USDA AgriculturalResearch Service laboratory in Beltsville, Maryland,with researcher and geneticist Dapeng Zhang, one ofthe world’s leading experts in cacao genetic diversity.

The FCIA has a rigorous submission protocol forcocoa beans being entered into the HCP program. It isdescribed in detail on the FCIA website (www.finechocolateindustry.org/hcp). Following conversion of thebeans into liquor and 68.1 percent cacao chocolate

using standardized procedures, the FCIA-HCP Sen-sory Panel renders an evaluation of the flavor. At thispoint, the sample is considered as HCP recommendedbut A site visit and genetic characterization arerequired before HCP status is awarded.

At the site we see if the fermentation process is astandard system, or if they’re adding something toadulterate or make the flavor different. Then we lookat their management practices. Are they doing some-thing really great that we need to know about, that wecan pass along in the future? Finally, we collect thegenetic material. We do not use any living material —we take a small, dried piece of a leaf and analyze theDNA within the leaf.

On the DNA side of it, we’re looking at somethingcalled single nucleotide polymorphic markers (snp ,pronounced “snip” for short). Every organism haschanges in their DNA, even humans. It’s as unique asyour fingerprint. You’ve got hundreds of thousands ofthese base pair changes within your DNA which youreceived from your parents and that make you unique.The same is true for cacao. We use a high-throughputanalysis system that can analyze thousands of thesemarkers at the same time. It is a very complicated sys-tem and it takes some time to do. After our site visit itmight take several months to get the samples to the lab,and a few more months after that to actually analyze itand run it through our process. Once we’ve identifiedthat tree or group of trees, we then compare the resultsto what we know about other populations that we’vealready looked at. We’re looking not only at the popu-lations, but we’re actually looking at their geneticmakeup and how they’re interrelated.

Figure 1 is our current DNA population map thatwe’re using now. It is constantly evolving because everysample that comes in is added to our library of infor-mation. There are numerous populations. You can seethat the morona/santiago population includes nacionalnow. We’re trying to encompass everything and bring itall together so we can understand how all of thesegroups are interrelated.

HCP DESIGNATE SAMPLES

The first four HCP-designate samples were awardedHCP status in January 2014 and unveiled at an FCIA

Anatomy of Fine Flavor

Pastaza, Nucuray, Ungumayo

San ago, Morona

Tigre

Beni (Bolivia)

Nanay, IMC

Amelonado

Madre de Dios Criollo

GU (French Guiana)

Napo

UcayaliNacional

Standard DNA Population Map

Figure 1


public event at the Institute of Culinary Education inNew York on February 11, 2014. The first two sampleswe’ll discuss are from Bolivia and, as you will see, theydiffer genetically and they also differ by location.

Bolivia Alto Beni

The Bolivia Alto Beni cocoa is grown and produced byVolker Lehmann. This is a population of trinitario clones(Figure 2). It falls within the classic trinitario group(crosses between criollo and amelonado) or it has somebit of influence of the upper Amazon. These are trinitar-ios that were moved into Bolivia decades ago andthey’ve been managed to such a point where they arerecognized now as heirloom cacao.

Now, what’s interesting about this is that you couldlook at this and say, “Oh trinitario, it’s halfway betweenthe amelonados and the criollos, and it’s got dispersiongoing up into the nanays and paranaris; I would expectcomplexity out of this. It’s covering a lot of genetic realestate here. I expect there to be things happening.”

It’s a little bit darker in color than the Tranquilidadsamples, with slightly orangish hues. The aroma ischocolate and fruit with green stem-like character. Thedepth of the chocolate flavor that comes out in thissample is striking. Remember the genetic positionbeing cut through between the trinitarios and the amel-onados, which is the basic source we know of today ofthis rich cocoa flavor. There are fruit and floral over-tones, slightly spicy, resinous with green notes and a

trace of vanilla character. The fruit overtone has citruswith slight balsamic wood resin and aged fruit notes.The chocolate presents a complex mix of chocolate,fresh citrus and aged fruits. This symphony of flavor hasa nice residual finish.

Bolivia Tranquilidad Estate

The second HCP-designated cocoa is also provided byVolker Lehmann but from the Tranquilidad Estate.This is a different location in Bolivia and represents anew population group. It’s not too far away from theAlto Beni but it is run by the same group, so it’s underthe same degree of management; they’re doing thesame kind of things with the postharvest processing ofthe beans. Nevertheless, you will find a different flavorprofile in the chocolate due to a combination of thegenetics and the microenvironment, the terroir.

The Tranquilidad samples are all falling within a newpopulation group from Bolivia and southern Peru (Fig-ure 3). This cacao type represents another group ofsemi-domesticated cacao like criollos. It’s inherentlyvery homogeneous, unique and it’s special from a fla-vor standpoint. All of the cocoas that we are tasting andwe have qualified as HCP designates are very complex.They are not simple chocolates.

We can note the rich, medium dark brown color andreddish hues in this sample. The aroma is a mildly richchocolate base, with fruit and tart aromatics. Thesebeans are on the more fermented side. That is what


BAB01BAB02

BAB03

BAB04

BAB05BAB06

BAB07BAB08 BAB09

BAB10BAB11

BAB12

BAB13

BAB14

BAB15

BAB16

BAB17

BAB18BAB19

BAB20

BAB21 BAB22

BAB23

BAB24

SQEC1SQEC2SQEC3SQEC4

SQEC5SQEC6

SQEC7

SQEC8

SQEC9SQEC10

SQEC11SQEC12SQEC13SQEC14

SQEC15

SQEC16

SQEC17

SQEC18SQEC19

SQEC20

SQEC21 SQEC22

SQEC23

SQEC24SQEC25SQEC26

SQEC27

SQEC28

SQEC29

SQEC30

SQEC31

SQEC32SQEC33

SQEC34

SQEC35

SQEC36

SQEC37

SQEC38

SQEC39

SQEC40

SQEC41SQEC42

SQEC43

SQEC44SQEC45

SQEC46

SQEC47

SQEC48

SQEC49SQEC50

EET 103

EET 183EET 19

EET 454

EET 48

EET 75

EET 95

EET 96

La Gloria 13

La Gloria 22

La Gloria 7

La Gloria 9Las Brisas 16

Las Brisas 30

Las Brisas 7

MO 118

MO 28

MO 82

IMC 10IMC 103

IMC 2IMC 33

IMC 67IMC 76

IMC 87 NA 286 NA 311

NA 399

NA 43

NA 79NA 916

PA 107PA 126

PA 191PA 30

PA 30

Amelonado 22

BE 10

IndioRed Amelonado

Amelonado 15

Criollo 13

Criollo 22 Criollo 12

Hon CriolloNIC Criollo

ICS 6ICS 40

ICS 5

ICS 83ICS 89 OC 77

Hacienda Limon Traquildad / Beni

Tranquilidad/Beni

Hacienda Limon

Bolivia Tranquilidad Estate DNA PrincipalCoordinate Plot

Figure 3

Upper Amazon

nAncie

forasteUpper Amazon Hroforaste iiawa A nio BetlA

Bolivia Alto Beni DNA Principal Coordinate Plot

Figure 2


drives some more of these dark notes, but there’s anunderlying brightness as well. It’s a very complex sam-ple, with a very interesting fruit character aftertastethat seems to remain in the mouth. It has a long finishin that sense. There’s an inherent sweetness associatedwith it. Sweetness is actually a part of the presentationof this bean and we don’t know why.

We have these two Bolivian samples that are differ-ent from one another. We can make some sense out ofthe genetics. But they’re also grown in a different loca-tion from one another physically. It could be the soil. Itcould be the climate, although these are probably a lit-tle closer together than many other places. The bottomline is, they’re different from one another.

Ecuador Hacienda Limon

From Bolivia we go to Ecuador. This sample was sentby Samuel von Rutte, at Orecao in Ecuador. His estateis the Hacienda Limon and it is located down the hillfrom the Vulcan Cotopaxi in the flat cocoa lands.

The genetics associated with this sample are very dif-ferent than the previous ones that we’ve talked about(Figure 4). This one lies within criollo and ancestralnacional or pure nacional. There is a very tight clusterrepresenting a very limited diversity within what he hasbut it is very unique. Mostly, von Rutte’s clones fallwithin a single group. Also, there are a few outside thecluster that are a little more associated with some ofthe other crosses. If we look at what he has, we see amodern variant of the classic, older nacional type.

Around the 1920s, this material was prevalent in mostof Ecuador and it was a trinitario crossed with nacional.He also has the amelonado/nacional cross. He has twobasic types that are linked together which form theuniqueness of his chocolate.

The Hacienda Limon cluster is very tight comparedto the spread of the Tranquilidad, Bolivia. One wouldexpect there to be some tightness or pointedness to thecharacter of the sample.

For an Ecuadorian this particular sample is surpris-ingly light in color, with reddish and distinct yellowishhues. Many people are familiar with the nacional arribasand know the depth of their color. The aroma is verymild, with chocolate, nut skin, nuts and mild red fruits.The initial flavor profile is mild but persistent. It hascomplex nutty and caramelic notes that dominate with asmooth background. This flavor was not produced as abyproduct of the processing. This is inherent in the beanand it happens every time we roast the beans andprocess them. What you see here is what came with thebeans; it’s what von Rutte did with his genetics.

Maunawili Experimental Station

The next sample we’re going to discuss is from Hawaii, onthe island of Oahu at the Maunawili experimental station.It is an experimental stand of trees that is owned by theHawaii Agriculture Research Center. It’s deliberately acollection. It never started out as a production planting.Daniel O’Doherty from Cacao Services went there, pulledthem all together and processed them into the cacao.


LCT EEN 163/ALCT EEN 241

LCT EEN 280

LCT EEN 302

LCT EEN 73/S

LCT EEN 83/SFortunato 4

MO 96

MO 99

MO 121

IMC 10

IMC 14IMC 20

IMC 31

IMC 33 IMC 50

IMC 55IMC 58

IMC 65

IMC 78IMC 94

IMC 96

NA 13

NA 40 NA 70

NA 79

NA 111 NA 399NA 670

NA 712

NA 807NA 824

PA 67

PA 194

OC 77

ICS 4

ICS 8

ICS 15

CHUAO 120

ICS 39 ICS 43

ICS 48ICS 60

ICS 63ICS 70

ICS 76

Amelonado15Amelonado22

SIAL325

Criollo13

Criollo 22

Amelonado(NIC)

KC1KC2C1

Hawaii Alto BeniUpper Amazon forastero

Ancien

Maunawili DNA Principal Coordinate Plot

Figure 5

Figure 4

Ecuador Hacienda Limon DNA PrincipalCoordinate Plot


You can see a cluster with a number of the Hawaiiansamples fitting really closely with the criollo (Figure 5).In this principal coordinate plot, the main take-away isthat the closer together something is, the more relatedit is. The others fall between the mixture of amelonado,your classic trinitario or are trinitarios with the addi-tion of upper Amazon forasteros. So it’s a very broadmix. It almost covers the entire map. It probably has avery complex flavor.

The Hacienda Limon was a little light for an Ecuado-rian. This Hawaiian sample is lighter yet, with slightreddish and yellowish hues. Think about the criolloblood in its genetics. It is very interesting that some ofthe trees are carrying that. There’s also a dispersion anda little bit more aroma in this sample, a mild chocolatearoma with ripe and slightly overripe fruit notes.There’s a lot of brightness in the flavor, with a mild,fresh-fruit acidity and an inherent mild spice note.

This sample is very soft and melts easily in the mouthas well. The beans also have a higher fat content. In theplant world, if you want something to be harder as a fat,meaning higher temperature melt point, go closer tothe equator to get it. If you want something to be asofter fat, meaning lower melt point, go north or south.For example, Bahia is about 15° south and is nearingthe lower southern limit of southern cocoa growing.Bahia cocoa butter was known, when it was available,as one of the softest cocoa butters that you could pur-chase. If you wanted a really hard butter you wouldlook for West African or Malaysian butter, meaninghigher-temperature melt point. It does not have any-thing to do with the heat. It is the daylight cycles thatthe plant experiences. The higher latitude in the north-ern hemisphere (or lower in the southern hemisphere)triggers in all plants the sequestration of the moreunsaturated fats, which makes the fat softer. Hawaii isactually above the 20° mark north at 21.5° and there-fore is getting a fair amount of day and night shiftingover the course of the year. This influences the plantand the plant then generates a softer fat. So you’veseen a difference created by the plant. This is clearly anenvironmental effect. To some extent, the observationthat the sample seems more fatty in the mouth is also amanifestation of the softer, more unsaturated cocoabutter because when the fat melts so easily you see it

more. You experience it longer in the mouth becauseit’s there as liquid fat longer.

There is a lot of diversity here. One of the challengeswhen you have that level of diversity is how exactlydoes one choose to optimize the fermentation? If Ihave something like porcelana, criollo, merediño orguasare, those beans ferment very quickly. Ideally, mostpeople say three days maximum. If you attempted toferment an amelonado for three days, you won’t getmuch chocolate flavor and it would be highly astrin-gent. It’s an interesting dilemma. So inherently, themanufacturer here had to play some compromisegames because he has that diversity in the population.So there are some things that you can readily deduce.Knowing that population structure, even though wedon’t know the exact genes that cause everything, wecan make some judgments or guesses on the connec-tions with flavor as well as other plant traits. You canmake judgments if you walk around the field and cutopen a lot of beans, but it’s nice knowing there’s actualscience behind it other than one’s eye.

Ecuador Coopertiva Nueva Esperanza

At this stage we’d like to unveil to you the fifth desig-nate for HCP status. It comes from the CoopertivaNueva Esperanza in Ecuador. It’s downhill from Quitoin the state of Pichincha which goes almost directlytowards the Pacific Ocean. It was provided by JoseMeza and Barbara Williams, the husband-and-wife


EET19

EET 48

EET 62

EET95

EET103

EET 400EET 544

Regulo 1

Regulo 2Mar 54

Fortunato 4

NA127

NA178NA191

NA232

NA26NA31 NA32

NA33

NA670NA702

NA710

NA756

NA916

POUND7

LCTEEN412

MO121

MO80

MO96

MO99

IMC2

IMC11IMC16

IMC31

IMC47IMC50

IMC54IMC60 IMC65

IMC67

IMC71IMC103

IMC105

DR1

ICS39

ICS41ICS43

ICS5ICS60

ICS68

ICS75

ICS76ICS84

ICS95

ICS97

OC77

Amelonado15

Amelonado 22

Catongo

Indio Rojo

SIAL 325

CRIOLLO13

CRIOLLO22

Finca Sr. Mesa

INIAP

Refractario

Nacional (pure)

Pound

Trinitario

Amelonado

Criollo

Ecuador Coopertiva Nueva Esperanza DNAPrincipal Coordinate Plot

Figure 6

team from Mindo Chocolate Makers in Michigan.We are very fortunate that we have worked with

Barbara almost two years ago. We had received someof her samples and they fall in large patterns. You’llnotice where the old nacionals are, the criollos are andthe amelonados are. Now let’s look at the structuremap of it (Figure 6).

You’ll see that they all fall within this same group.There’s nacional influence, criollo influence and amel-onado influence. It’s in every single one of them. Now,there’s another group that matches up very closely withthem. These are called the refractario clones. The veryold nacional from the 1920s were collected by F. J.Pound when he went there looking for witches’ broom-resistant nacional clones. A lot of this material is sittingin Trinidad and we’ve looked at it numerous times aspart of our reference collection. We can see that a lot ofthese match up with what Mindo has at the site. It’svery nice, very old, classic nacional types.

Every single one of these vertical bars is an individ-ual tree (Figure 7). So the diversity within this popula-tion in terms of the parentage (that is, the familycontribution structure) is very similar even thoughgenetically with the snp markers it spreads across afairly wide range. The reality of it is that the parentagerepresents a very specific, defined base. Yet each indi-vidual in this population is unique and genetically dif-ferent, just like every single human is unique.

Our fifth HCP designate is also a medium brown

with reddish and yellowish hues, but the color is darkercompared to the color of the Hacienda Limon, alsofrom Ecuador. It has a rich chocolate aroma with a hintof wood. There is a little bit of the caramel nut notesthat we found in the Hacienda Limon, but nowherenear the level. There is more chocolate flavor. We’regetting some of the more classic, arriba-like notes,although not quite as floral as you would expect to seein say, manabi, vinces or esmeraldas beans at times.Mild fruit acidity and astringent notes continue to drivethe nut/nut skins character along with a plum/darkfruit/complex fruit note with a trace of tobacco spiceand very mild green floral notes. This complexity differ-entiates this sample from the classic arriba cocoas. Wesuspect that some of this complexity is due to the diver-sity and range of those genetic markers.

We learned from the site visit, too, that these treesoriginated from seeds. When you go back to that struc-tural chart, you’ll see that there’s a lot of genetic diver-sity that’s very similar to children. You have shifts inthe crossing between the parents.

Here is a good example. Looking again at Figure 7, anindividual bar that has a lot more of the green color,which represents the nacional, is going to be a differentindividual in its character and nature from a sample thathas a much smaller level of that particular contribution.Genes that influence the nature and characteristics of aparticular cacao are derived from their parentage.

CONCLUSION AND FUTURE DIRECTION

That broad map showing the entire world of cacao as weknow it today is continuing to be expanded. We expectthat the HCP samples will continue to drive some of thatexpansion. Ultimately, our objective is to understandwhat about the genome leads to the kind of functionalchanges that occur in fermentation and in drying and thatthen yield flavor when we process it. We do not knowhow all of these variables are interconnected today butwe are adding to the body of knowledge that will let usdo that at some point in time. At the end of the day, themission of the HCP is to ensure that the diversity of fla-vor that exists within the world called Theobroma cacaois preserved for generations to come. n



Finca Sr. Meza

Refractario (Nacional hybrids) Pure Nacional Nanay

Morona IMC (Iquitos) Trinitario

FA-1

FA-2

FA-3

FA-6

FA-7

FA-1

3FA

-16

FA-1

7FA

-24

FA-2

5FA

-26

FA-2

8FA

-30

FA-3

1FA

-32

FA-3

4FA

-35

FA-4

0FA

-41

FA-4

2FA

-43

FA-4

5FA

-48

FA-4

9FA

-51

FA-5

2FA

-53

FA-5

4FA

-60

FA-6

1FA

-151

FA-1

52FA

-153

FA-1

54FA

-155

FA-1

56FA

-157

FA-1

58FA

-159

FA-1

60FA

-163

FA-1

64FA

-166

FA-1

70FA

-171

FA-1

73FA

-174

FA-1

75FA

-176

FA-1

78

EET1

9EE

T48

EET6

2EE

T95

EET1

03EE

T400

EET5

44AM

1/57

AM1/

8AM

1/95

B12/

1B1

2/2

B5/7

B6/3

B7/1

4B9

/10-

25CL

10/1

5CL

10/2

7CL

19/1

0EE

T400

JA1/

19JA

5/25

LP1/

41LP

4/32

LP4/

8LV

20LV

28LX

31LX

43M

OQ6/

19SJ

1/19

SJ1/

40SL

C18

CCN

51Re

gulo

1Re

gulo

2M

ar-5

4Fo

rtuna

do4

NA12

7NA

178

NA19

1NA

232

NA26

NA31

NA32

NA33

NA67

0NA

702

NA71

0NA

756

MO1

21M

O80

MO9

6M

O99

IMC2

IMC1

1IM

C16

IMC3

1IM

C47

IMC5

0IM

C54

IMC6

0IM

C65

IMC6

7IM

C71

IMC1

03IM

C105

DR1

ICS3

9IC

S41

ICS4

3IC

S5IC

S60

ICS6

8IC

S75

ICS7

6IC

S84

ICS9

5IC

S97

OC77

Amel

onad

o15

Amel

onad

o22

Cato

ngo

Indi

o Ro

joSI

AL32

5CR

IOLL

O13

CRIO

LLO2

2

Amelonado Criollo

Ecuador Coopertiva Nueva Esperanza GeneticMarkers

Figure 7Presented at the FCIA summer event

cacao flavor through genetics—anatomy of fine flavor · 2014. 12. 2. · 26 november 2014 † the...

Documents