“Potato taint” in African coffees
Some East African coffees, especially those from Rwanda, but also Burundi, Tanzania, Zambia, and Kenya, are afflicted by the strong flavor of potato peels, not a desirable taste in coffee. This defect is said to be ultimately caused by one of the pyrazine chemical compounds. How do coffee beans end up with this chemical? The prevailing theory is that coffee cherries are damaged by insects, most notably several species of stink bugs. The damage facilitates contamination of the cherry, which leads to formation of the potato taint compounds.
The connection between the bugs and the taint are not completely understood. Let’s take a super-geeky look at what we know about the bugs — which, whether they cause the potato defect or not, do a lot of damage to coffee cherries — and how they might be responsible for the potato taint.
Bugs that make a stink
The insects that cause damage to coffee cherries in East Africa that are usually associated with potato taint are in the order Hemiptera, the true bugs. This order includes various bugs that suck plant juices, including cicadas, leafhoppers, aphids, scale insects, shield or stink bugs, and many others. The stink bugs that are coffee pests are collectively called “Antestia bugs,” after their former generic name (many species in the genus Antestia have been reclassified under another genus, Antestiopsis). Two primary culprits are Antestiopsis orbitalis (formerly Antestia lineaticollis) and Antestiopsis intricata.
A. orbitalis (above right) is a colorful bug about 7 mm long. Various subspecies have slightly different patterns. This and related species that also attack coffee have similar life histories; for the sake of simplicity I’ll refer to them collectively as variegated coffee bugs. Some species are found in India and southeast Asia. We’ll focus on the ones usually found in Africa.
Variegated coffee bugs can complete four generations a year, and reproduce best at temperatures between 19 and 24 degrees C (66-75 F) and humidity between 35-50%. Eggs are laid on the undersides of leaves. The young are called nymphs, and are similar, but smaller, than the adults. Average life span is three to four months. Like all Hemipterans, variegated coffee bugs have piercing mouthparts adapted to sucking plant juices. Arabica coffee (rarely robusta) is the preferred host, but other plants in the coffee family (Rubiaceae) are also used. Variegated coffee bugs feed on shoots and leaves (causing damage and bud drop) but primarily on unripe coffee cherries.
Not only does this type of feeding itself cause physical damage to the cherry, but fungi (yeast) in the genus Nematospora (N. [=Eremothecium] coryli and N. [=Ashbya] gossypii) can secondarily infect the cherry. Nematospora fungi are not specific to coffee, but when these two species infect coffee, it’s usually called coffee bean rot. The fungi only cause rot in unripe (green) cherries.
It’s believed that the bugs are vectors of the fungi; that is,the bugs carry the spores and the fungi are dependent on the bugs (though not variegated coffee bugs exclusively) for dispersal. There is some dispute over whether the spores are present internally in the bugs and thus inoculated into plants, or if they are present on the surface of the bugs. Not all cherries pierced by the bugs become infected. It could be that the fungal spores are not present in/on all bugs, or that if they are present internally, they may be too large to pass through the mouthparts of younger (smaller) individuals.
Physical damage to the coffee cherry and the associated rot cause significant losses (up to 45%) on coffee farms infested with variegated coffee bugs. Do these bugs also cause potato taint?
Chemicals that make a stink
A number of chemical compounds produce potato-like odors. The most notable is a methoxypyrazine: 2-methoxy 3-isopropylpyrazine (or “MIPP”*). The odor threshold of MIPP is very low, so it’s easily detected in very small quantities. MIPP has been found in nature in some plants and higher organisms. Various pyrazines can be synthesized chemically and biologically, and MIPP has been produced by cultures of at least one bacteria, Pseudomonas perolens.
Other Pseudomonas cultures also have potato odors. This group of bacteria are free-living, and widely found in soil and water (at least one strain infects the leaves of coffee trees). Several other bacteria, such as some strains of Serratia and Cedecea, produce potato-like odors that are the result of a combination of pyrazine compounds.
Despite my access to vast quantities of scientific literature, I was surprised to find virtually no published research on the chemical processes of coffee bean rot. Does it produce MIPP or a similar compound with a potato odor? If other bacteria are involved in the potato defect, are they connected in some way to the fungal infection that is typically introduced by variegated coffee bugs? Where to the bacteria come from?
One very interesting clue comes from Tim Hill, of Counter Culture Coffee, who provided the photo above left. He said that the potato odor was apparent in the air during a rainstorm in Burundi. This is suggestive that a/the taint-producing bacteria may be present in the soil. While I have been unable to pin down the range of Pseudomonas perolens, there are nearly 200 species of this bacteria worldwide. I have to wonder why the potato defect is largely (exclusively?) considered an East African problem and why it has been historically linked to variegated potato bugs, but not, for instance, coffee berry borers which also penetrate the green cherry. The borers and Pseudomonas are fairly ubiquitous in coffee-growing nations. It seems to me that there must be a link between the variegated coffee bugs and a bacteria that facilitate the production of stinky pyrazine compounds.
Bringing us to this compelling clue: MIPP (usually going by its synonym IPMP*), is found in some grapes and contributes to pleasant flavors in wines in small amounts, but at higher levels is associated with the off-flavor known as “ladybug taint.” The ladybugs (Asian multicolored lady beetles, Harmonia axyridis, the non-native species that can be a household pest) do not actually attack or harm the grapes. IPMP is part of the chemical make-up of the ladybugs, and when the insects get mixed in and processed with the grapes, the taint occurs in the wine.
IPMP is present in lots of ladybug species, and many other insects that are “aposematic” — those possessing some kind of warning signal to potential predators. Usually, this is some sort of bright coloration, very often red and black. I have not seen any variegated coffee bugs or close relatives on lists of insects that have been confirmed to have any pyrazines, but their colorful patterns are consistent with other aposematic insects, and some other Hemiptera are classified as aposematic. Recall this group of bugs is known as “stink bugs.” This is precisely because most have the ability to release a nasty chemical when molested. So further exploration of the chemical make-up of variegated stink bugs surely seems a promising avenue of research.
However these compounds end up in the coffee cherry, they end up altering the bean, which itself does not show damage. (This fact — that the damaged cherries must be identified and discarded prior to processing, after which they cannot be detected until the coffee is roasted or ground — is what makes this defect so frustrating.)
Recent news out of the University of California, Riverside announced that one of their entomologists was going to Rwanda to help solve the mystery of the potato defect. That item said, “there is no definitive link between potato taste and antestia bug, only hypotheses.” While the research I’ve cited (see below) is not very current, the dots seem to be connected right up to the end point of why and how MIPP or a similar compound is produced.
The battle of the bug
Given the fact that one way or another, variegated coffee bugs are pests of coffee, control methods for them will continue to be important. Fungicides do not control the type of infection caused by Nematospora, given that the fungi are introduced within the coffee cherry. Small infestations of the bugs can been battled with hand-picking. Since the bugs like dense foliage, pruning is often recommended. In the long run, both natural and synthetic pyrethrum insecticides have proven ineffective in many cases. The bugs have typically been controlled with multiple applications of pesticides, usually fenitrothion, chlorpyrifos, malathion, trichlorfon, and diflubenzuron. All but the last are organophosphate pesticides that are especially dangerous (to humans and the environment) when not applied according to instructions with full protection, which is often not the case in less-developed nations.
Fortunately, because they are native to East Africa, variegated coffee bugs do have many natural enemies which may be exploited for biocontrol; they are especially vulnerable to a number of native parasitic wasps that attack the eggs. With persistence and luck, reliable biological and cultural control of variegated coffee bugs will hopefully be developed.
As the Rwandan and Burundian specialty coffee sectors grow, the urgency to defeat the potato taint will grow. I’ll be following any progress and research on the exact mechanisms of potato taint and any methods of control and detection that emerge.
Photo of Antestiopsis orbitalis by Lambert Smith, used with permission.
*This compound has several synonyms: 2-Isopropyl-3-methoxypyrazine, 3-Isopropyl-2-methoxypyrazine, or IPMP. The CAS Registry number is 25773-40-4.
Coffeed forum thread, “The Rwandan Potato problem,” June 2007.
James Hoffman, “The Phantom Potato,” February 2009 blog post. This post and the comments, and the forum post above talk about potato taint from the barista/consumer perspective in particular. Below, is some of the primary academic and scientific literature.
- Cheng, T.-B., G. A. Reineccius, J. A. Bjorklund, and E. Leete. 1991. Biosynthesis of 2-methoxy-3-isopropylpyrazine in Pseudomonas perolens. J. Agric. Food Chem. 39:1009-1012.
- Cilas, C., B. Bouyjou, and B. Decazy. 1998. Frequency and distribution of Antestiopsis orbitalis Westwood (Hem., Pentatomidae) in coffee plantations in Burundi: implications for sampling techniques. Journal of Applied Entomology. 122:601-606.
- Crowe, T.J., G.D.G. Jones, and R. Williamson. 1961. The use of pyrethrum formulations to control Antestiopsis on coffee in East Africa. Bulletin of Entomological Research. 52:31-41.
- Greathead, D.J. 1966. A taxonomic study of the species of Antestiopsis (Hemipteea, Pentatomidae) associated with Coffea arabica in Africa. Bulletin of Entomological Research. 56:515-554.
- Kirkpatrick, T. W. 1937. Studies on the ecology of coffee plantations in East Africa. Ii. the autecology of Antestia Spp. (pentatomidae) with a particular account of a Strepsipterous parasite. Transactions of the Royal Entomological Society of London 86:247-343.
- Le Pelley, R.H. 1932. On the control of Antestia Lineaticollis, Stål (Hem., Pentatom.) on Coffee in Kenya Colony. 1932. Bulletin of Entomological Research. 23:217-228.
- Le Pelley, R.H. 1942. The food and feeding habits of Antestia in Kenya. Bulletin of Entomological Research. 33:71-89.
- McNutt, D.N. 1979. Control of Antestiopsis spp. on coffee in Uganda. Tropical Pest Management. 25:5-15.
- Mehrotra, R. S., and Aggarwhal, A. 2003. Plant Pathology, 2nd Ed. Tata McGraw-Hill, New Delhi.
- van der Meulen, H.J., and A.S. Schoeman. 1990. Aspects of the phenology and ecology of the antestia stink bug, Antestiopsis orbitalis orbitalis (Hemiptera: Pentatomidae), a pest of coffee. Phytophylactica. 22:423-426.
- Mitchell, P.L. 2004. Heteroptera as vectors of plant pathogens. Neotropical Entomology. 33:519-545.
- Nixon, G.E.J. 1941. New Braconid parasites of Antestia Lineaticollis, Stål, and of Sylepta Derogata, F. Bulletin of Entomological Research. 32:93-101.
- Pickering, G. J, M. Spink, Y. Kotseridis, D. Inglis, I. D. Brindle, M. Sears, and A. Beh. 2008. Yeast strain affects 3-isopropyl-2-methoxypyrazine concentration and sensory profile in Cabernet Sauvignon wine. Australian Journal of Grape and Wine Research. 14:230-237.