Dangerous herbs

DNA Barcoding and traditional Chinese medicine

Chinese herbs have been used as medicine for over 2000 years. There are some 13,000 natural medicinals used in China with over 100,000 recipes recorded in the ancient literature. More and more these traditional drugs are used outside of China and some producers of Chinese herbal medicines are pursuing FDA clinical trials to market their products as drugs in U.S. and European markets. Therefore, the authentication of the ingredients is becoming a critical, international issue because mistakes can cause illness and even death. Plant elements and extracts are by far the most common elements used and there is confusion and substitution happening. Not a surprise given the large number of potential ingredients for these products.

A couple of years ago a 60 year-old man from Hong-Kong was diagnosed with kidney failure and cancer of the urinary tract. He had been taking an herbal prescription and the case was investigated by the local Department of Health. The investigations revealed that the patient was mistakenly given a product called Herba Aristolochiae Mollissimae (made from Aristolochia mollissima) instead of Herba Solani Lyrati (made from Solanum lyratum). Herba Aristolochiae Mollissimae is known to contain the poisonous aristolochic acid which indeed can cause kidney failure and is known to be cancerogen when taken over a longer period of time.

Aristolochia mollissima

Solanum lyratum

In the wild both plants can be distinguished rather easily (see pictures) but some characteristics may vary with growing stages, environmental properties, physical forms, and post-harvest processing of the herbs. Confusion is further facilitated by the fact that both herbs share the same Chinese common name Bai Mao Teng (meaning  ‘white hair-bearing vine’).

Now a group of researchers from Hong Kong tested if DNA barcoding and chemical fingerprinting are useful alternatives to the use of microscopy and elaborate chemical analysis. Indeed they are. The chemical profiling was able to detect the harmful component aristolochic acid. However, the chemical composition of a herb can vary between different life stages and can be altered during processing. In contrast, DNA barcoding  is not affected by these factors and usually a small amount of sample is sufficient for DNA extraction. The study also showed that the smallest DNA fragment amplified was about 50 bp long which demonstrates that even partially degraded DNA could be retrieved. The researchers tested a variety of markers (the standards matK and rbcL, as well as ITS, trnH-psbA and trnL-trnF) which provides us with an arsenal of possible tests that will help to avoid such fatal confusions.

An unconventional field site

Scientists from the Florida museum published the first study on butterflies and moths of Guantanamo Bay Naval Station and discovered large biodiversity in an area previously unknown to researchers and one you wouldn't normally think of when it comes to plan field work. Appearing hopefully soon in the Bulletin of the Allyn Museum, the study creates a baseline for understanding how different plant and animal species have spread throughout the Caribbean.

Guantanamo Bay Naval Station

 

One of the authors, Jacqueline Miller says, "Biodiversity studies are extremely important because they give us clues about where things were and how they evolved over time so we can better understand what may happen in the future.We're also looking at climate change over time, and butterflies are biological indicator species since they are associated with particular plants as caterpillars and often found in particular habitats."

In January, researchers collected 1,100 specimens representing 192 moth and 41 butterfly species, including the invasive lime swallowtail whose proximity to the U.S. poses a threat to citrus plants. The researchers also froze tissue samples from many of the collected specimens for future DNA Barcoding analysis.

Leased to the United States in 1903 (although this is disputed by Cuba), the land has unintentionally become a wildlife refuge, offering researchers the opportunity to better understand the island's natural habitats. "Because it is a military base -- and this is true for many military bases, which typically have large areas of land -- people are not trampling, bulldozing or developing the land," says Roger Portell, another author of the study. "So there is a large area of land in the southeast corner of the island that has basically been untouched for 100 years."

Indeed such pristine environments can often be found on many military bases or similarly protected areas. A good example for the latter is the former border separating West and East Germany. It was estimated that around 600 threatened species of animals and plants were given a free rein in a no man’s land overshadowed by minefields, metal fences and watchtowers. Today the former border is turned into a unique chain of nature reserves running for nearly 1,400km in a gentle zigzag from the Vogtland region, near the German-Czech border in the south, to the Baltic Sea in the north, to form what is called the green belt.

This could serve as a good example what to do with such areas once the military has given them up. I haven't given up the hope that this happens to Guantanamo Bay one day.

Standards

A DNA Barcode is a short standardized sequence enabling species discrimination

This short definition contains all important elements of DNA Barcoding and this post is about one very important word in it: standardized

Standards are paramount in our work and it is concerning that an increasing number of studies use different gene regions and their authors call it DNA Barcoding. I find it this even more puzzling when it comes to vertebrates where the reasons for not using COI are either traditional or even worse - the result of convenience. 

It is always exciting when new species are discovered by using DNA methods and I fully support the idea that any phylogenetic approach needs additional genetic markers, the same is true if a sufficient placement can't be made using DNA Barcoding alone. However, I will never be able to compare my barcode data on fishes with data of colleagues who decided to use cyt b instead. This is already frustrating when you try to build phylogenies. 

Phylogenetics is a good example what can go wrong when you don't agree on standards. In essence everybody can use whatever marker they think works best for them. Fortunately, that started to change over the last few years. Well, we are living in a free world, and I am the last who is going to tell somebody else how to do their job but the result is that most data that is publicly available isn't necessarily comparable. With years of experience in trying to assemble datasets for particular taxonomic groups in order to build robust phylogenies I consider it a waste of money and ignorant not to use data other researchers have generated before. However, often I simply can't do that as the allowable amount of unknown data in an analysis is limited. There isn't sufficient overlap between datasets as communities of researchers haven't agreed on common gene regions for this kind of analysis. On the contrary there were long lasting disputes about which region(s) to use. Some battles still continue.

For me DNA Barcoding represented a big leap forward by proposing standard regions (and features) and in some cases it took longer find a consensus among all scientists (e.g. the plant barcode) but they did. Nevertheless, currently we have standard regions for 3 major groups: 

• Cytochrome Oxidase I (COI or COXI) for animals
• Ribulose-bisphosphate carboxylase gene (rbcL) and the chloroplast maturase K gene (matK) for plants
• The complete ITS1 spacer, the 5.8S gene, and the ITS2 spacer as a single contiguous sequence (ITS) for fungi

There might be compelling reasons (especially technical ones) not to use one of those in the particular groups and people are of course always entitled to chose whatever works best for them but I would like to make two suggestions:

(1) Use the term DNA Barcoding only if you use one of the markers listed above to indicate that you follow agreed upon community standards.

(2) Give the markers at least a try to ensure that you've done what you can to contribute to a global effort and the community of your colleagues.

More on DNA quality

A few posts ago I was talking about a new publication that put DNA Barcoding sequences to some quality tests. Today the journal MycoKeys published a paper that represents a compilation of best practices which can help with the quality management of sequence data. 

The lead author of the paper, Henrik Nilsson, on the motivation to write the publication: "Many researchers find sequence quality control difficult, though. There just isn't any straightforward document to put in their hands to give them a flying start. As a result, scientists differ in the degree to which they are aware of the need to exercise sequence quality control and in what measures they take."

The authors focus on the DNA Barcode for fungi, ITS but all suggestions and recommendations are broadly applicable to other markers and organisms. The paper is a collection of really good guidelines and especially useful for starters. This is one of the papers I would give any freshman student to read if I had any :-)

Here a summary of their 5 guidelines (Table 1 of their publication).

Target of guideline	Way of getting there
1. Establish that the sequences come from the intended gene or marker	Do a multiple alignment of the sequences and verify that they all feature some suitable, conserved sub-region (here the 5.8S gene)
2. Establish that all sequences are given in the correct (5’ to 3’) orientation	Examine the alignment for any sequences that do not align at all to the others; re-orient these; re-run the alignment step; and examine them again
3. Establish that there are no (bad cases of) chimeras in the dataset	Run the sequences through BLAST in INSD/UNITE and verify that the best match comprises more or less the full length of the query sequences
4. Establish that there are no other major technical errors in the sequences	Examine the BLAST results carefully, particularly the graphical overview and the pairwise alignment, for anomalies
5. Establish that any taxonomic annotations given to the sequences make sense	Examine the BLAST hit list to see that the species names produced make sense

Thanks!

I would like to take the opportunity to thank everyone for the support in the first week of my blog. I was very surprised about the extremely positive response and I am very proud to have reached 500 unique page views on some days, and that I already have 11 followers, not counting all RSS feeds :-)

This is very encouraging and motivating and I will continue with my goal of a post per day (maybe not on weekends though - family time comes first). There is certainly enough out there that deserves a post. I might miss things but I strive to be fast (my "CSI: Rainforest" was the second entry on that news in the web - not bad for a journalism rookie). If you have something interesting that you think might deserve a post e.g. something you are currently working on and are about to publish please let me know. I am happy to beat the press once more :-)

Please feel free to post comments. Feedback is always welcome. Blogs are usually very short (for good reasons) and commenting is a way to add information to a post or discuss the topic as such. I am all up for that. For now I haven't put many restrictions on that function as my site is currently far from being popular enough to attract spam or comments of strange people. Of course an ambitious blogger strives for many more readers but I am realistic enough not to expect to be able to compete with the big science blogs. Nevertheless, tell your family, friends, colleagues, students, sponsors... there is a blog on DNA Barcoding not only for scientists.

Thanks a lot.

Quality Test for DNA Barcodes

A new publication appeared in PLoS ONE proposes a direct way of assessing sequence errors in published records.

The authors tested the hypothesis that sequencing errors in reference barcodes can be detected as very low frequency variants at sequence positions that are otherwise highly conserved. They used their approach to assess sequencing error in a large dataset of bird sequences (11,333 sequences from 2706 species) which they obtained at GenBank. They used both sequences with the keyword "BARCODE" and such without. The keyword indicates that the DNA Barcodes follow the so called barcode-standard with a minimum of 500 bp from a defined region (COI in animals), linkage to museum specimens, and publicly archived trace files documenting a minimum quality score.

I think this study shows three very interesting results:

Prevalence of VLFs over time (Stoeckle & Kerr 2012)

(1) The very low frequency variants (VLF) detected in single individuals of a species were mostly concentrated at the ends of the barcode sequence, consistent with sequencing error. The maximum error rate is estimated at approximately 0.05 errors per barcode sequence.

(2) The method was able to recognize a number of overlooked cryptic pseudogenes lacking stop codons which are usually the best indicator for those.

(3) The high overall quality of the dataset especially in comparison with COI sequences deposited before implementation of the barcode standard thereby supporting its effectiveness. 

A very nice paper with most analytics done in Excel - Can't wait to have a look at some of my fish datasets.

CSI: Rainforest

Brazilian police officers have traveled all the way to the Royal Botanical Garden in Edinburgh, Scotland to receive training in DNA Barcoding at Pete Hollingsworth's lab. They are forensic experts who want to extend their expertise to put an end to illegal logging in the Amazonian rainforest.

Brazilwood (Credit Encyclopedia Britannica)

Illegal logging is on the ­increase in Brazil, which is home to a large number of rare and protected tree species. Criminal gangs working in the Amazon rainforest fell the trees to satisfy a black market for the wood both in Brazil and abroad. It is a lucrative business for the loggers, e.g. the wood for a single violin bow is worth several thousand dollars. Just recently the police confiscated several violins which are believed to be made from brazilwood (Caesalpinia echinata) an endangered tree that gave Brazil its name. However, the currently used methods are not sufficient to provide a conclusive result and therefore it is very difficult to get a conviction. 

Violin bows made from Brazilwood

DNA Barcoding might be exactly what is needed to prove that seized wood or wood products are from protected species and there is great interest to develop a barcode reference library of plants that are protected from international trade. It is about time to equip law enforcement authorities with a tool that provides conclusive evidence in such cases. It is almost unbelievable but it is estimated that illegal logging alone causes losses in assets and revenue in excess of 10 billion Dollar annually.