Spatial Distribution of Collections Yielding Marine Natural Products

Abstract

The societal benefits of coral reef ecosystems include shoreline protection, habitat provision for reef fish, tourism, and recreation. Rarely considered in valuation of reefs is the considerable contribution of marine natural products (MNP) to both human health and the economy. To better understand the relation of MNP discovery with the characteristics and condition of coral reef ecosystems, we initiated a study to track the collection location and taxonomic identity of organisms that have provided pharmacological products. We reviewed collection information and associated data from 298 pharmacological products originating from marine biota during the past 47 years. The products were developed from 232 different marine species representing 15 phyla, and the 1,296 collections of these specimens occurred across 69 countries and seven continents. Our evaluation of the collection data was hampered by sundry observational and reporting issues, including imprecise location descriptions and omission of collection dates. Nonetheless, the study provides an important synopsis and appraisal of years of study and exploration by the marine natural product community. Understanding and quantifying the benefits of MNP discovery will depend upon improved reporting of collections, including accurate taxonomic identification, collection dates, and locations.

INTRODUCTION

Coral reef ecosystems are widely seen to provide numerous valuable ecosystem goods and services (EGS). Until recently, those EGS had not been sufficiently cataloged and quantified to be included in policy and decision-making assessments. To remedy this gap, the U.S. Environmental Protection Agency (EPA) prepared a series of reports that described and quantified these benefits to the extent possible. In preparing these reports, we observed that the pharmacological potential of marine natural products (MNPs) was rarely (if ever) included in policy or decision support documents, probably because assessors are generally unaware of the potential and no metric or predictive model exists for estimating that potential. Importantly, such a metric or model might be based on the condition or state of the ecosystem being assessed. To determine whether a plausible basis existed for creating such a pharmacological potential metric or model, we first had to establish where and when the specimens were collected that yielded MNPs with pharmacological potential (for which we used clinical status as a surrogate). This review provides that necessary foundation for future investigations seeking a predictor of pharmacological potential.

This perspective, the spatial distribution of collections, does not appear to have garnered much attention in the literature. There have been a few country-focused reviews (see below) and some work by Leal and colleagues¹ to estimate trends in the discovery of new MNPs at the fairly aggregated levels of Exclusive Economic Zones (EEZ), Large Marine Ecosystems (LME), and Marine Ecosystems of the World (MEOW).² Mukherjee and colleagues³ demonstrated the potential of depicting collection data in a geospatial system such as Google Earth, but their emphasis was on demonstrating the available tools and datasets.

Reviews related to MNPs tend to fall into two broad categories: (1) temporally based reviews that appear annually or biennially or cover a multi-year period; and, (2) topically based reviews that focus on a chemical family, a taxonomic group, a therapeutic use, a country, or some other categorical basis. Probably the best example of the first category are Faulkner’s annual reviews of newly isolated MNPs of pharmacological interest that began in 1984 and continued until his untimely death in 2002.⁴ Blunt and colleagues have continued this annual review.⁵ From 2003 to 2013, Hill published an annual review of the highlights of the MNP literature organized around the MNPs’ chemical structures and that focused on synthetic aspects.⁶ Mayer and colleagues published annual and biennial reviews of preclinical research on MNPs covering the period from 1998 to 2011, usually in two parts: one covering MNPs demonstrating anticancer potential and a second covering MNPs showing potential for all other therapeutic uses.⁷ Butler and colleagues have published periodic reviews of MNPs in clinical testing covering from the end of 2004 to the end of 2015.⁸ There are several examples of reviews that cover 20 to 30 years: Faulkner published two such reviews⁹ and Cragg and Newman and colleagues have published several such reviews¹⁰ that cover both marine and terrestrial natural products, as well as numerous useful reviews of natural products in various stages of development.¹¹ While these temporal reviews rarely provide collection data (such as locations and collection dates), they are a rich source of information about chemical structures, pharmacological effects, clinical trials, efficacy, ownership, patents, and the primary sources that do report collection data.

Topical reviews covering MNPs are numerous in both quantity and variety. Most common are reviews that focus on a chemical family,¹² a phylum,¹³ or a therapeutic use.¹⁴ Less common reviews focus on a genus or family,¹⁵ a country,¹⁶ or biodiversity.¹⁷ As with the temporal reviews, we used these topical reviews mainly to identify primary sources of collection data as well as to provide information about chemical structures, pharmacological effects, clinical trials, efficacy, and ownership.

Our approach was to first identify MNPs that had received Food and Drug Administration (FDA) approval for therapeutic use or that reached the clinical testing stage of development. We assumed that clinical status was a reasonable surrogate for pharmacological potential (i.e., the likelihood that the MNP would be approved by the FDA, and in turn yield market and nonmarket economic benefits). We later added MNPs identified in the literature as in preclinical testing or as promising leads. The goal was not to have a representative sample of chemical structures or geographic locations, but rather to identify locations yielding the MNPs with demonstrated value or potential value. Then, we identified the species from which those MNPs had been isolated and the locations where the specimens yielding those MNPs had been collected. We report here our findings to date and the data on which they are based, covering 298 MNPs and 1,296 specimen collections (Table 1). The spatial distributions of those collections are summarized in several tables. In the process of compiling the collection data, we were surprised to find that many articles reported incomplete, vague, or inaccurate location descriptions or collection data, which would impede any attempt to replicate the collections and could adversely affect the accuracy and usefulness of any metric or model based on those data. We briefly discuss some examples of these uncertainties in the hope that investigators will be encouraged to report accurate and complete collection data. Finally, we describe potential lines of future investigation that could lead to the development of a metric or model for estimating a location’s of pharmacological potential, and we suggest the need to establish a process for maintaining the data.

Table 1.

Overall Statistics for the Compiled Data

MNPs reviewed	298
Species collected	232
Phyla represented	15
Specimen collections documented	1,296
Countries where collections occurred	69
Continents where collections occurred	7
Years in which specimens collected	47
Clinical status of MNPs
Approved by the FDA	16
In clinical trials	55
In preclinical testing	51
Lead or probe	176

DESCRIPTION OF THE SUPPORTING INFORMATION DOCUMENTS

The compiled data and analyses are available in the Supporting Information (SI) documents (Table 2). The Ancillary Tables document (SI#1) contains three tables that are only referenced in this article: a list of the MNPs having incomplete data; an expanded list of review articles; and, a table of examples of the problems we encountered in reports of collections. The Overview (SI#2) has three chapters: Chapter 1 is a brief review of the origins of this research; Chapter 2 has background information for the five ensuing documents (SI#3-SI#7); and, Chapter 3 has extensive summary tables and figures of the compiled data. The Marine Pharmaceuticals Table (MPT), the Species Collections and Distributions Table (SCT), and the Gazetteer repeat the document-specific background information from the Overview’s Chapter 2. Note that the two worksheets of the Collections Database (CDc and CDs) contain identical data, but the CDc is organized by the country of collection, while the CDs is organized by the species collected (Table 2). The reference standards we used are discussed immediately below, followed by a synopsis of each of the Supporting Information documents (except Ancillary Tables), beginning with the MPT (SI#3) then continuing seriatim to cover the SCT, the Collections Database, the Gazetteer, and the Google Earth Spatial Database of Collections (GEDB) (SI#7) before turning to the Overview (SI#2) for several summary tables of the collection data. This ordering approximates the workflow of compiling the data.

Table 2.

Supporting Information Documents (n.b.: the five PDFs are combined in a single file)

Code	Document Name	Type	Document Contents
SI#1	Ancillary Tables	PDF	Three tables cited only in this article: MNPs with incomplete data; MNP reviews bibliography; and, collection reporting problems
SI#2	Overview and Summary Tables (Overview)	PDF	background information on how we processed the data and summary tables and figures of the data collected
SI#3	Marine Pharmaceuticals Table (MPT)	PDF	a table of MNPs with their clinical status, chemical class, source species, therapeutic uses, and source references
SI#4	Species Collections and Distributions Table (SCT)	PDF	a table of source species listing their phylum, distribution, and collections; collection source citations listed in table
SI#5	Collections Database (by Country [CDc] & by Species [CDs])	XLS	a spreadsheet database with one sheet sorted by country and a second sheet sorted by species; each collection has an entry on each sheet; see Table 4 for data list; the CDc and CDs have identical data
SI#6	Gazetteer of Collection Geography (Gazetteer)	PDF	a table of relevant sites listing the BGN Approved Names and Name Variants and lat-lon coordinates in both DMS and decimal notation
SI#7	Google Earth Spatial Database of Collections (GEDB)	KMZ	a spatial database of collection sites using the Google Earth application; placemark icons show the degree of locational certainty

REFERENCE STANDARDS

Given the diversity of sources and locations, we believed it prudent to rely on several reference standards, as described below. It should be noted that even reference standards can create interpretive challenges and cannot be followed blindly.

World Register of Marine Species (WoRMS).¹⁸

WoRMS provided binomial names, taxonomic characterizations, and species’ distributions. WoRMS is inherently mutable, because it tracks changes in genus/species names and assignments. Given that the highest priority for WoRMS is taxonomic tracking, the species distributions listed in WoRMS may be incomplete.¹⁹

GeoNet Names Server (GNS) & Geographic Names Information System (GNIS).²⁰

Reference standards for location names and geographic coordinates were provided by two databases, both being official repositories of standard spellings specified by the U. S. Board on Geographic Names (BGN): (1) the GeoNet Names Server (GNS) maintained by the U. S. National Geospatial-Intelligence Agency (NGA); and, (2) the Geographic Names Information System (GNIS) maintained by the U. S. Geological Survey (USGS). The GNIS contains location data for sites in the United States, its dependent areas (e.g., Guam, Puerto Rico), and Antarctica, while the GNS contains similar data for all other countries. When collection sources used a name or spelling other than an Approved Name, the reported name or spelling is provided in the Notes field. The Gazetteer (SI#6), contains entries for the approved names as well as cross-references for commonly used name variants and other names reported in collection sources.

Japanese Geographic Suffixes.

These suffixes can be confusing or inconsistent in translation. For example, the Japanese suffix -wan (Kanji character ) can mean either “bay” or “gulf’, so Sagami-wan has been reported as both “Sagami Bay” and “Gulf of Sagami”. While “Gulf of Sagami” may refer to the area further offshore (closer to the Pacific Ocean), that area is sometimes referred to as Sagami-nada, where the suffix -nada (Kanji character ) can mean either “gulf’ or “sea”. The GNS is not consistent in its treatment of these suffixes. Many approved names (e.g., Iwate-ken, Ago-wan, Kuro-shima, etc.) have the suffix hyphenated and in lower case. However, some approved names (e.g., Aomori Ken, Sendai Wan, Mage Shima, etc.) capitalize and do not hyphenate the suffix. For consistency and to minimize confusion, we have used the hyphenated-lowercase format in all tables and databases; however, in the Gazetteer, the entry for the approved name(s) is in the form reported by the GNS. A list of Japanese language geographic suffixes can be found in the front matter of the Overview, the MPT, the SCT, and the Gazetteer, as can a list of non-English geographic terms found in placenames.

Country, Province, & State Codes.

Country name abbreviations are normally the three-letter country code used by the International Olympic Committee (IOC).²¹ These codes were chosen over the ISO 3166–1 alpha-3 country codes,²² because the IOC codes tend to be more evocative of the country name (hence, easier to remember). In a very few instances where the ISO code better evokes the country name, that code is used rather than the IOC code. When there is no IOC country code, the ISO code is used. When neither an IOC code or an ISO code is available, we created a three-letter code. The abbreviations used for provinces and states in Australia, Canada, and the United States are the two-letter or three-letter postal abbreviations. A table listing the abbreviations used can be found in the Overview (Table SI#2–2), the SCT (Table SI#4–2), and the Gazetteer (Table SI#6–2).

SI#3 MARINE PHARMACEUTICALS TABLE (MPT)

The Marine Pharmaceuticals Table (MPT) lists the 298 MNPs (or MNP families [e.g., brevetoxins]) shown in Table 3 that were the focus of our analyses. The type of data found in the MPT is shown in Table 4. Also listed in the MPT are 363 MNPs or MNP families (Table SI#1–1) that may have pharmacological potential, but for which we have only partial information at this time. All of the sources cited in the MPT can be found in the bibliography following it.

Table 3.

Marine Natural Products Reviewed

aaptamines	chlorovulone	ω-conotoxin CnVIIA	ethyl icosapentatea
AB-5	chondroitin sulfate	ω-conotoxin CVIDb	eunicidiol
abeohyousterone	chondropsins	ω-conotoxin MVIIAc	fascaplysin
abyssomicin C	ciliatamides	contignasterold	floxuridinee
acyclovif	clavulone	contulakin-Gg	fluorouracilh
adociasulfate-2i	cometins	Thr10-contulakin-Gj	FR290581k
agelasphins	conantokin-Gl	cortistatins	frondoside A
agosterol A	conantokin-Tm	cryptophycins	fucoxanthin
amphimedine	α-conotoxin AnIA	curacins	GA3 polysacchariden
anabaseine	α-conotoxin AnIB	cytarabine ocfosfateo	geodiamolides
aplidinep	α-conotoxin AnIC	Dacinostatq	geoditins
aplyronine A	α-conotoxin ArIA	dehydrothyrsiferols	girodazole
Ara-Ar	α-conotoxin ArIB	deoxyamphimedine	girolline
Ara-Cs	α-conotoxin CnIA	1 –deoxynojirimycint	glembatumumab vedotinu
arenastatin A	α-conotoxin CnIB	deoxyvariolin Bv	globostellatic acids
arenjimycin	α-conotoxin EpI	dermatan sulfatesw	globostelletins
arenosclerins A–C	α-conotoxin ImI	diazepinomicinx	halichondramide
ascidiacyclamide	α-conotoxin PnIA	diazonamide A	halichondrin B
ascidideminy	α-conotoxin PnIB	dichlorolissoclimide	haliclonacyclamine E
ASG-5ME	α-conotoxin Vc1.1z	dictyodendrins A–J	(–)-haliclonadiamine
aurilides	α-conotoxin cVc1.1aa	dictyostatin-1	haligramides
avarol	α4/7-conotoxin Lp1.1	didemnin B	halimidebb
axinelloside A	α4/7-conotoxin TxIA	dideoxypetrosynol	haouamine A
barbamide	αA-conotoxin PIVA	dihydromotuporamine	haouamine B
bastadins	γ-conotoxin TxVIIA	discodermins	haterumaimides
bengamides	δ-conotoxin PVIA	discodermolide	hemiasterlincc
bengazoles	δ-conotoxin TxVIA	DMMC	herbadysidolides
bistramides	ε-conotoxin TxIX	DMXBAdd	homophymines
bistratamides	ϰ-conotoxin PVIIAee	dolastatin 10	(+)-8-hydroxymanzamine A
bistratenes	ϰA-conotoxin CcTx	dolastatin 15	(–)-8-hydroxymanzamine A
brentuximab vedotinff	ϰA-conotoxin SIVA	dysidenins	hyousterones A-D
brevetoxins	ϰI(2)-conotoxin ViTx	dysidins	ianthelliformisamines
bromovulone	μ-conotoxin CnIIIB	dysiherbaines	iodovulone
bryostatin 1	μ-conotoxin CnIIICgg	eilatin	IPL-512602hh
bursatellanin-P	μ-conotoxin SIIIAii	elcatoninjj	IPL-550260
callipeltins	μO-conotoxin MrVIBkk	eleutherobin	IPL-576092ll
callystatin A	ϱ-conotoxin TIA	elisidepsinmm	isofagominenn
Cemadotinoo	χ-conotoxin MrIApp	ER-076349	jaspamideqq
cephalostatins	χ-conotoxin MrIBrr	ER-086527	jaspolides
chlorolissoclimide	ψ-conotoxin PIIIE	eribulin mesylatess	jorumycin
kahalalide F	miglustattt	phakellistatins	stelliferins
kahalalide O	mirabalins	philinopside A	stolonic acids
keenamide A	mirabamides	pibocin B	stolonoxides
kottamides A–E	mirabilins	Plinabulinuu	stylisins
KRN-7000vv	mollamides	poecillastrins	styloguanidinesww
LAF-389	monomethylauristatin Exx	polybrominated diphenyl ethersyy	superstolides
lamellarins	monomethylauristatin Fzz	psammaplins	swinholides
largazole	motuporamines	psammaplysins	Taltobulinaaa
latrunculins	mycalamides	R-135853	tamandarin A
laulimalidesbbb	mycalosides	reidispongiolides	tamandarin B
laurenditerpenol	mycothiazole	renieramycins	Tasidotinccc
lissoclibadin 1	namenamicin	rhabdastrellic acid–A	tawicyclamides
lissoclibadin 2	neoamphimedine	rhabdastrellins	tetrodotoxinddd
lissoclinamides	neodysiherbaines	rhabdastrins	thiocoraline
lissoclinolides	neosiphoniamolides	(–)-salicylihalamides	Trabectedineee
lissoclinotoxin	neosurugatoxin	saliniquinones A–F	trodusqueminefff
lophotoxin	Neovastatggg	sansalvamide	trunkamide A
Lovazahhh	norsegoline	sarcodictyin A	tuberatolide A
luffariellins	okadaic acid	sarcodictyin B	tuberatolide B
luffariellolides	onnamides	saxitoxin	ulicyclamide
lurbinectediniii	(–)-palau’amine	SGN-75	ulithiacyclamides
makaluvamines	panobinostatjjj	shishijimicins	variolin Bkkk
manoalides	papuamides	soblidotinlll	viequeamides
manzamine	(–)-papuamine	sphinxolides	vitilevuamide
marizomibmmm	patellamides	spirodysin	wakayin
meridianins	patellazoles	spisulosinennn	Xen-2174ooo
methopterosinppp	patellins	Spongistatinsqqq	Zalypsisrrr
methyleudistomins	peloruside A	Squalaminesss	zidovudinettt
miglitoluuu	peptidolipins	stelletins

Unless otherwise noted, these are synonyms:

^aEPA-E, EPA ethyl ester, Epadel, ethyl eicosapentaenoic acid.

^bAM336, CNSB004, leconotide.

^cAMM336, C1002, SNX-III, Prialt, ziconotide.

^dIZP-94005.

^eFUDR.

^fZovirax; made by total synthesis with MNP pharmacophore.

^gCGX-1160.

^h5-FU, Adrucil, Efudex, Fluroplex. There is some doubt that fluorouracil is an MNP, given the rarity of fluorinated natural products and a lack of replication of the isolation by Xu et al., plus that a biosynthesis pathway to add fluorine to pyrimidines has not been discovered.

ⁱAS-2.

^jCGX-1063.

^kAS-1710313.

^lCGX-1007.

^mCGX-100.

ⁿGA3P.

^ofosteabine, SPAC, Starasid, YNK-01; sometimes misspelled Starsaid.

^pAplidin, dehydrodidemnin B, DDB, plitidepsin.

^qNVP-LAQ824.

^rarabinofuranosyladenine, adenine arabinoside, vidarabine, Vira-A; see f.

^sarabinosyl cytosine, cytosine arabinoside, cytarabine, Cytosar U, DepoCyt, Tarabine PFS; see f.

^tAT2200, moranoline.

^uCDX-011, CR011-vcMMAE.

^vPM01218.

^wKRX-101, sulodexide.

^xEC0–4601, TLN-4601.

^ysometimes spelled ascididemnin.

^zACV1.

^aaIMB-007.

^bbNPI-2350, (−)-phenylahistin.

^ccE7974.

^ddDMXB-anabaseine, GTS-21; sometimes misspelled DMBXA.

^eeCGX-1051.

^ffAdcetris, SGN-35.

^ggXEP-018.

^hhAVE 0547.

ⁱⁱPEG-SIIIA.

^jjCalcimar, carbocalcitonin, Miacalcin.

^kkCGX-1002.

^llHMR-4011A.

^mmIrvalec, PM02734.

ⁿⁿAT2101, IFG, Plicera.

^ooLU-103793; sometimes misspelled cematodin.

^ppCMrVIB, χ-conopeptides, χ-CTX MrIA, mr10a.

^qqjasplakinolide.

^rrχ-conopeptide.

^ssE7389, Halaven, norhalichondrin B.

^ttOGT 918, Zavesca.

^uuNPI-2358.

^vvagelasphin, α-galactosylceramide, α-GalCer.

^wwisopalau’amines.

^xxMMAE.

^yyPBDE.

^zzMMAF.

^aaaHTI-286.

^bbbfijianolides.

^cccILX651, synthatodin.

^dddTectin, tetrodin, TTX.

^eeeecteinascidin 743, ET-743, Yondelis.

^fffMSI-1436.

^gggӔ-941.

^hhhω-3-acid ethyl esters.

ⁱⁱⁱPM01183.

^jjjFarydak [ex- Faridak], LBH-589.

^kkkPM-01220.

^lllauristatin PE, TZT-1027.

^mmmNPI-0052, salinosporamide A.

ⁿⁿⁿES-285.

^oooχ-conopeptide.

^ppp0AS-1000, pseudopterosin A methyl ether.

^qqqaltohyrtins, cinachyrolides.

^rrrPM00104.

^sssEvizon, MSI-1256F.

^tttAZT, Retrovir; see f.

^uuuDiastabol, Glyset.

Table 4.

Contents of Major Tables

No.	SI#3 MPT Columns	SI#4 SCT Columns	SI#5 CDc / CDs Columns	SI#6 Gazetteer Columns
1	Chemicala	Species & AphialD	Collection Country	Country
2	Status Category	Phylum & MNPs	CDc ID	Site or Feature
3	Chemical Class	Distribution	CDs ID	Feature Type
4	Source Species	Collection Locationsb	Collection Location	BGN Approved Names
5	Therapeutic Uses	Notes	Species	BGN Name V ariants
6	Clinical Status Details		Phylum	Other Names
7	Source References		AphiaID	GNS/GNIS lat-lon DMS
8			Chemical	GNS/GNIS lat-lon decimal
9			Clinical Status	Unique Feature Identifier
10			Collection Source	Description & Notes
11			Source Page
12			Collection Date
13			Year Collected
14			Month Collected
15			Year Rcvd by Journal
16			Year Published
17			Depth of Collection
18			Reported Location Lat
19			Reported Location Lon
20			Placemark Decimal Lat
21			Placemark Decimal Lon
22			Placemark Lat/Lon Source
23			Placemark Location Name
24			Placemark Label
25			Icon
26			Icon in Words
27			Double Entries
28			Notes

^aincludes synonyms and sponsors.

^bincludes source reference, collection date, and collection depth

Names of MNPs.

The MNPs in the MPT are most often identified by their common (or trivial) names, while their semisystematic, testing, and trademarked names are usually listed as synonyms. For example, the common name panobinostat is used for sorting, while the testing name (LBH-589) and the trademark name (Farydak) are reported as synonyms (see also Table SI#3–3). The use of trivial names, while convenient, presents some offsetting pitfalls. Most common are misspellings (e.g., ascididemin vs. ascididemnin; cemadotin vs. cematodin; DMXBA vs. DMBXA; Staracid vs. Starsaid). In several instances, there have been competing names for a compound when different research groups working independently isolate and name the same compound (e.g., irciniastatin vs. psymberin; laulimalides vs. fijianolides; altohyrtins vs. cinachyrolides vs. spongistatins). In one instance, the same name was given to different MNPs, so one had to be renamed (mirabalin and mirabilin).

Therapeutic Uses of MNPs.

These uses are limited to those we found in the literature. Although a large number of bioassays are reported in this table, few specimens have been thoroughly tested in the whole range of possible bioassays. Because almost all of the specimen collections were conducted by academic researchers with limited resources, virtually all of the specimens have only been tested in bioassays for cancer cytotoxicity. Consequently, their potential use in treating cardiovascular diseases, central nervous system disorders, diabetes, infections (e.g., HIV, tuberculosis, parasites, malaria, etc.), asthma, and immunological disorders remains largely unexplored.²³

Clinical Status of MNPs.

The 298 MNPs are categorized according to their clinical status: (1) Lead or Probe; (2) Preclinical Testing; (3) Clinical Testing; or, (4) Approved. Except for approved drugs, determining the clinical status of an MNP can be difficult, because pharmaceutical companies tend to closely hold such information. Similarly, determining the chemical’s owner/controller/sponsor can be quite complicated.

SI#4 SPECIES COLLECTIONS AND DISTRIBUTIONS TABLE (SCT)

The Species Collections Table (SCT) lists the 232 marine species (or genera) (Table 5) from which the 298 MNPs were isolated. The type of data found in the SCT is shown in Table 4. All of the source references can be found in the bibliography that follows the SCT.

Table 5.

Collections by Species

Aaptos aaptos	137	Conus anemone	2	Ecteinascidia turbinata	25
Aaptos ciliata	1	Conus arenatus	15	Elysia ornata	2
Aaptos suberitoides	7	Conus catus	12	Elysia rufescens	3
Aaptos sp.	5	Conus consors	2	Erythropodium caribaeorum	7
Acanthaster planci	6	Conus episcopatus	8	Eudistoma bituminis	1
Acanthostrongylophora sp.	6	Conus geographus	7	Eudistoma gilboviride	2
Agelas mauritiana	7	Conus imperialis	12	Eudistoma sp.	4
Amphimedon sp.	2	Conus leopardus	14	Eunicea fusca	1
Amphiporus angulatus	1	Conus magus	5	Eunicella cavolini	2
Amphiporus lactifloreus	1	Conus marmoreus	13	Fusarium sp.	2
Antillogorgia elisabethaea	5	Conus pennaceus	7	Geodia corticostylifera	5
Aplidium albicans	1	Conus purpurascens	1	Geodia gibberosa	1
Aplidium haouarianum	1	Conus striatus	16	Geodia sp.	2
Aplidium meridianum	1	Conus textile	15	Gymnodinium A3	1
Aplysia kurodai	3	Conus tulipa	6	Halichondria (Halichondria) sp.	1
Aplysinella rhax	3	Conus victoriae	1	Halichondria melanadocia	1
Aplysinella strongylata	1	Conus virgo	5	Halichondria okadai	2
Aplysinella sp.	1	Coriocella hibyae	1	Halichondria sp.	1
Arenosclera brasiliensis	9	Cucumaria frondosa	4	Haliclona (Reniera) sp.	3
Auletta sp.	1	Cymbastela cantharella	4	Haliclona nigra	1
Axinella corrugata	2	Cymbastela sp.	6	Haliclona sp.	12
Axinella brevistyla	1	Cystodytes dellechiajei	4	Halimeda copiosa	6
Axinella infundibula	1	Cystodytes sp.	1	Halimeda lacrimosa	1
Axinella sp.	3	Dendrilla cactos	3	Halocynthia pyriformis	1
Babylonia japonica	1	Diazona angulata	1	Hapalochlaena maculosa	2
Bellonella albiflora	1	Dictyodendrilla verongiformis	1	Hemiasterella minor	1
Botryllus tuberatus	1	Didemnum chartaceum	2	Histodermella sp.	1
Bryopsis pennata	1	Didemnum cuculliferum	2	Holothuria (Halodeima) grisea	1
Bugula neritina	18	Didemnum fragile	1	Homophymia sp.	1
Bursatella leachii	5	Didemnum molle	6	Hymeniacidon sp.	7
Cacospongia mycofijiensis	2	Didemnum obscurum	2	Hyrtios altus	2
Callipelta sp.	2	Didemnum sp.	5	Hyrtios communis	1
Callyspongia truncata	1	Discodermia dissoluta	1	Hyrtios erectus	4
Calthropella (Calthropella) sp.	1	Discodermia kiiensis	1	Ianthella basta	14
Cephalodiscus gilchristi	2	Discodermia sp.	3	Ianthella flabelliformis	2
Chondropsis sp.	1	Dolabella auricularia	29	Ianthella quadrangulata	2
Cinachyra sp.	1	Dysidea arenaria	2	Ianthella reticulata	1
Clathria sp.	1	Dysidea avara	2	Ianthella sp.	1
Clavelina sp.	1			Ircinia ramosa	4
Clavularia viridis	1			Ircinia sp.	3
Colochirus quadrangularis	1
Jaspis digonoxea	1	Neopetrosia exigua	2	Spongia sp.	6
Jaspis johnstoni	8	Neopetrosia sp.	1	Squalus acanthias	1
Jaspis serpentina	1	Neosiphonia superstes	2	Stelletta clavosa	1
Jaspis splendens	6	Niphates caycedoi	2	Stelletta sp.	4
Jaspis sp.	5	Ochrophytab	—	Stolonica socialis	3
Jorunna funebris	3	Pachypellina sp.	1	Stolonica sp.	1
Kirkpatrickia variolosa	5	Paranemertes peregrina	1	Styela clava	1
Lamellaria sp.	1	Parasphaerasclera albiflora	2	Styela plicata	2
Lamellodysidea herbacea	61	Parasphaerasclera aurea	1	Stylissa caribica	1
Laurencia intricata	1	Petrosia (Petrosia) ficiformis	1	Stylissa carteri	8
Laurencia viridis	3	Petrosia sp.	4	Stylissa constricta	1
Lendenfeldia chondrodes	6	Phakellia costata	4	Stylissa massa	8
Leptoclinides sp.	3	Phakellia fusca	3	Suberea ianthelliformis	5
Lissoclinum badium	3	Phakellia sp.	5	Suberea sp.	1
Lissoclinum bistratum	8	Phallusia nigra	4	Suberites domuncula	1
Lissoclinum patella	33	Pleurobranchus albiguttatus	1	Suberites sp.	2
Lissoclinum timorense	2	Pleurobranchus forskalii	2	Symploca hydnoides	11
Lissoclinum sp.	8	Poecillastra sp.	1	Symploca sp.	1
Lissodendoryx sp.	4	Polysyncraton lithostrotum	2	Synoicum adareanum	2
Lophogorgia sp.	10	Prorocentrum belizeanum	1	Takifugu chrysops	4
Luffariella variabilis	60	Prorocentrum concavum	1	Takifugu niphobles	2
Luffariella sp.	6	Prorocentrum lima	10	Takifugu oblongus	16
Mactromeris polynyma	1	Psammoclema sp.	1	Takifugu pardalis	6
Melophlus sp.	1	Pseudoceratina purpurea	10	Tectitethya crypta	2
Micromonospora aurantiaca	1	Pycnoclavella kottae	1	Theonella mirabilis	3
Micromonospora marina	1	Raspailia (Clathriodendron) arbuscula	2	Theonella swinhoei	42
Micromonospora sp.	2	Raspailia (Parasyringella) elegans	1	Theonella sp.	9
Microscleroderma sp.	3	Rhabdastrella distinca	1	Thorecta reticulata	1
Monanchora arbuscula	4	Rhabdastrella globostellata	35	Trachycladus spinispirulifer	8
Monanchora unguiculata	4	Rhaphoxya sp.	1	Trididemnum cyanophorum	2
Moorea producensc	64	Rivularia sp.	1	Trididemnum cyclops	1
Mycale (Arenochalina) laxissima	2	Salinispora arenicola	20	Trididemnum orbiculatum	1
Mycale (Arenochalina) mirabilis	1	Salinispora tropica	7	Trididemnum solidum	6
Mycale hentscheli	21	Sarcodictyon roseum	1	Verrucosispora maris	9
Negombata magnifica	24	Siliquariaspongia mirabilis	2	Xestospongia sp.	4
Neopetrosia carbonaria	5	Siliquariaspongia sp.	1	Zopfiella marina	1
Neopetrosia contignata	1	Siphonochalina sp.	1	Zyzzya fuliginosa	6
		Smenospongia aurea	1

^aThe genus Pseudopterogorgia was transferred to Antillogorgia in 2012. See Table SI#4–3 for other such transfers. Also see SCT (SI#4).

^bFucoxanthin is found in most or all species of the classes Phaeophyceae (brown algae), Chrysophyceae (golden algae), and Bacillariophyceae (diatoms) in the phylum Ochrophyta plus some species in the phyla Dinoflagellata and Haptophyta, possibly as many as 16,000 species in total. These species and their collections are not included in the analyses presented here. See fucoxanthin entry in the MPT (SI#4) for references.

^cThe genus Moorea, proposed in 2012, includes many specimens attributed to Lyngbya majuscula. See Table SI#4–3 and SCT (SI#4).

Species Names.

The taxonomic identification of species can be complicated, particularly when reporting collections from 30–50 years ago, coupled with the dynamic nature of taxonomic classification. For example, in the past 25 years, sponges identified as Acanthella carteri, Acanthella aurantiaca, Axinella carteri, and Phakellia carteri have all been transferred to Stylissa carteri. WoRMS did not have an entry for Phakellia carteri until August 2014. Table SI#4–3 lists examples of name changes and taxonomic reassignments related to the 232 species. In the SCT itself, superseded names are given with the prefix “ex-”. The superseded names are often more common in the literature than the newer names. The “ex-” prefix is also used in other situations, such as when an MNP’s owner or sponsor changes.

Species Distributions.

The distributions of the species and genera listed in the SCT come from the WoRMS Taxon Details web page for the species (or genus). In some instances, the Distribution Details popup links provide more specific locational detail. We did not include locations marked “inaccurate” by WoRMS. Choosing among the various names for locations or geounits to include in the distributions involves numerous choices and assumptions, which are described in Section 1.2.7 of the SCT. Aside from certain Japanese names, the locations listed in the distributions represent the only exception to our use of BGN Approved Names. The distributions in the SCT use the names reported by WoRMS to minimize confusion when comparing locations in the SCT with those on the WoRMS Taxon Details web pages.

Unfortunately, for some species there seems to be little or no overlap between the distribution locations and the collection locations. This situation could arise because the distribution is incomplete or because the specimen was misidentified. The sponge Aaptos aaptos is as an example (Table SI#4–7). WoRMS reports that Aaptos aaptos is confined to European waters, but its distribution listing identifies numerous locations (all marked “inaccurate”) outside of European waters where Aaptos aaptos specimens were supposedly collected. Interestingly, all of the collection locations we report in the SCT are also outside of European waters. A Species Overview linked to on the WoRMS Taxon Detail webpage for Aaptos aaptos states that the species “has been reported from many areas of the world, but it is almost certain that these concern other species of Aaptos.”²⁴ So, with respect to identifying the species from which MNPs have been isolated, it is unclear whether the collected specimens were Aaptos aaptos, a single new Aaptos species, or several new Aaptos species, and the matter will remain unresolved until comparative taxonomic examinations of the voucher samples are conducted. In this and similar instances, the SCT and the Collections Database use the identification reported by the investigators, unless a more recent identification has superseded the original.

Specimen Collections.

The collection locations within a species (or genus) are listed in chronological order of the collection date, with two exceptions: (1) when no collection date is reported, the publication date of the source reference is used for ordering; and, (2) when a species has a very large number of collections, the collections are first grouped by country, then the countries are ordered chronologically by earliest collection, and then collections are ordered chronologically within countries. The publication date is often a poor approximation of the collection date (frozen preserved specimens may be analyzed 10 or more years after being collected), but it seemed to be the best surrogate available. The placenames of the collection locations are the BGN Approved Names (except for certain Japanese names; see Reference Standards supra). Reported placenames that are not the BGN Approved Name are provided in a note. Although collection depths are commonly reported as negative numbers, indicating distance below sea level, the SCT and the Collections Database report these values without the negative signs, primarily so that depth ranges are clearer.

Problem Resolution.

To the greatest extent possible, we endeavored to provide additional information for context and to explain our resolution of data discrepancies and conflicts in the Notes columns of the SCT, the Collections Database, and the Gazetteer (Table 4). The types of problems we addressed in this fashion fall into five broad categories and multiple subcategories (Table 6). Usually when there were two or more possible collection locations, we added a double entry so that both the reported location and our surmised location are represented. In the Collections Database, the double entry is noted in a separate column (column 27) and explained in the Notes column (Table 4). In the GEDB Spatial Database, these doubled locations are both displayed with a question mark icon (Table SI#2–15).

Table 6.

Types of Collection Reporting Problems Encountered

Type	Subcategory	Description
Collection Uncertainty
Collection Date Uncertainty		The collection date(s) is(are) not reported or reported in vague or nonspecific terms in the source reference. If the voucher specimen number is reported, it may be possible to infer an approximate collection date.
Collection Depth Uncertainty		The collection depth is described in qualitative terms.
Same Collection or Separate Collections?		It is uncertain whether specimens came from the same collection or separate collections.
Lat-Lon Errors
Arcminutes or Arcseconds Value Exceeds 60		Neither arcminutes nor arcseconds values can exceed 60, but occasionally larger values are reported in source references and must be interpreted.
Lat-Lon Transcription Error		An error is likely to have been made in transcribing the lat-lon coordinates.
Simplified Lat-Lon		The lat-lon coordinates point to an unlikely location probably due to rounding or simplifying the lat-lon values.
Location Uncertainty
Contradictory Location Description		The location description contains contradictory information, such as naming the collection location and the station number, but the two are at different sites.
Incorrect Directional Description		The reported directions describing the collection location are incorrect or incomplete.
Location Name Change		Subsequent to publication, the name of the collection location is changed.
Multiple Locations with Identical Name		Multiple locations in the same region or country have the same name, and it is unclear which was the location of the collection.
Location A Very Large Area		The reported location is a very large area (e.g., Australia). When a specific location is used to represent the collection location, an explanation is provided in the Notes field.
Unknown Location		The named collection location is not shown on maps and does not appear in the usual geographic resources.
Name Uncertainty
Name Ambiguity		The collection location may have multiple names from different source references, or the reported name may not be found in the usual geographic resources.
Name/Lat-Lon Mismatch		The reported collection location and the reported lat-lon coordinates do not coincide.
Non-BGN Approved Name		The placename reported for the collection location is not the BGN Approved Name; the reported placename is provided in a numbered note.
Specimen Identification
Binomial Name Change		Subsequent to publication, the specimen’s taxonomic classification changed; the reported binomial name is provided in a numbered note along with contextual information about the accepted name.
Binomial Name Error		The reported binomial name for the specimen is incorrect; the reported binomial name is provided in a note along with contextual information about the accepted name.
Binomial Name Uncertainty		The specimen’s correct binomial name cannot be ascertained with certainty.

When a series of collections are made, source references rarely, if ever, report specific collection dates. For example, a source reference might report that collections were made every two weeks from January to June without specifying start and end dates; or, that collections were made from July to December without specifying the number of collections or their frequency; or, that collections were made at several locations over several years. These situations required that we surmise the collection dates, and we provide our reasoning in estimating the collection dates in the Notes column.

SI#5 Collections Database (by Country [CDc] and by Species [CDs])

The collections database consists of two worksheets with identical data, one sorted by country of collection and the second sorted by the binomial name of the species collected. The two worksheets contain much of the collection-related data in the MPT, the SCT, and the GEDB (Table 2). The data for each of the 1,296 reported collections (plus 36 duplicate entries due to locational uncertainty) are contained in a row with 28 columns (Table 4). All of the references listed in the Collection Source column and the Notes column can be found in the SCT bibliography. In the Collections Database, we do not use numbered notes, because they would conflict with some of the operations of the spreadsheet. Instead, cells for which there is an explanatory note have bolded text and yellow fill.

SI#6 GAZETTEER OF COLLECTION GEOGRAPHY

While compiling these collections, we discovered that there are many instances where reported location names did not coincide with names found on maps or in geographic references. This gazetteer provides the geographic data (Table 4) for 1,365 collection locations and nearby sites and locales. To minimize confusion, we have used the BGN Approved Names for all collections, except as noted above for certain Japanese names.

SI#7 GOOGLE EARTH SPATIAL DATABASE OF COLLECTIONS (GEDB)

The GEDB was created using the data in the Collections Database and can be viewed using the Google Earth application. When viewed in Google Earth, the GEDB provides a sense of the spatial distribution of the 1,296 collection sites for all 232 species. Figure 1 is an example of the GEDB showing the Great Barrier Reef near Townsville, Queensland, Australia.

Figure 1.

An Example of the Google Earth Database Showing Collections Made From the Great Barrier Reef near Townsville, Queensland, Australia.

Placemark Icons.

Each unique species-collection location is represented by an icon (or placemark). Because multiple collections were made at some sites, the 1,296 collections (and 36 duplicate entries) are represented by just over 800 placemarks. The icon used for the placemark indicates the degree of locational certainty we associated with the collection (Table SI#2–15). For example, when latitude and longitude coordinates are reported by the investigators, the icon is a bullseye, but when we had to surmise a location based on a name alone, the icon is a circle. However, when the location named is a large area, such as an entire country, the icon is a diamond. For those instances when we had great uncertainty (usually due to misreporting of latitude or longitude coordinates), the icon is a question mark. The icons used are from Google Earth’s default icon set, so the GEDB can be used on any platform with minimal configuration issues.

Placemark Labels.

The placemark labels are mostly straightforward in content: the placename, the country code, and the binomial name of the species collected. Two fairly common additions are the notation {ctr} when the icon is a diamond (i.e, the placemark represents a large area) and the notation {n}, where “n” is the number of collections of the named species at the named location. An example placemark label having these elements is: Okinawa-ken (JPN) {ctr} {2} [Acanthaster planci]²⁵ When the reported collection location does not match the reported latitude-longitude coordinates, question mark icons are placed at both locations, and the notation “{actual}” added to the reported location placemark label and the notation “{reported}” added to the reported latitude-longitude placemark label. Labels with yellow text indicate that additional information can be viewed by double-clicking the placemark icon.

SI#2 OVERVIEW AND SUMMARY TABLES

The Overview describes the form, content, and methods of each of the six followingSupporting Information documents and provides 19 summary statistical tables and 12 figures to describe the data therein.

Collections by Country.

The first perspective taken for the collections is a “by country” comparison. It should be noted that not all of the “countries” we have used for organizational purposes are, in fact, countries (Table 7). Except for Antarctica, which is a bit of an anomaly, the non-countries are used because they provide a better sense of the collection location than the proper country (e.g., Cura9ao vs. the Netherlands, and New Caledonia vs. France). For the purposes of our analyses, all of these entities are referred to as countries, regardless of their actual political status.

Table 7.

Locations That Are Not Sovereign Countries

Location	Status
Antarctica	a de facto condominium under the Antarctic Treaty System
Bermuda	a British Overseas Territory
British Virgin Islands	a British Overseas Territory
Curaçao	a constituent country of the Kingdom of the Netherlands
French Polynesia	an overseas collectivity of France
Guadeloupe	an overseas region of France
Guam	a territory of the United States
New Caledonia	a special collectivity of France
Paracel Islands	disputed sovereignty between China, the Philippines, and Taiwan
Puerto Rico	a territory of the United States
South Georgia	a British Overseas Territory
United States Virgin Islands	a territory of the United States

There are also locations that are included within other locations. The Coral Sea Islands are included in Australia, as is the Torres Strait. Île Europa and Mayotte are included in France, and the Îles Chesterfield are included in New Caledonia. Both the Balearic Islands and Canary Islands are included in Spain, and the Palmyra Atoll is included in the USA. Many seas and gulfs that are adjacent to multiple countries are categorized under Oceans (e.g., the Gulf of Aqaba).

The collections by country entries shown in Table 8 are not certain values, because many source references are unclear regarding the number or frequency of collections. As a result, these values are our best estimate of the number of collections based on the descriptions provided by the investigators. In each such instance, we explain our estimation method in the Notes fields of the SCT and the Collections Database (Table 4). In addition, these values should not be taken as suggesting the relative size of different countries’ pharmacological potential from MNPs. Many different factors influence the choice of a locale to explore for MNPs, including the overall diversity of reef life, known populations of specific species, accessibility of the reefs, established working relationships with local institutions and vendors, cost of collecting specimens (including travel, equipment rental, contract services, etc.), national policies on collecting, ease of obtaining collection permits, etc. Consequently, the relative number of collections among countries seems most likely to result from observational bias (looking where it is most convenient, sometimes referred to as the “streetlight effect”).²⁶

Table 8.

Collections by Country and Clinical Status

	Collections	Approved	Clinical	Preclinical	Lead
Antarctica	7	—	—	5	2
Australia	147	3	66	11	67
The Bahamas	21	1	16	1	3
Bangladesh	3	—	3	—	—
Belize	13	6	2	2	3
Bermuda	1	—	1	—	—
Brazil	33	—	4	3	26
British Virgin Islands	1	—	—	—	1
Cambodia	1	—	1	—	—
Canada	3	—	—	3	—
China	6	—	—	—	6
Colombia	3	2	1	—	—
Comoros	6	4	2	—	—
Croatia	1	—	—	—	1
Cuba	1	—	—	—	1
Cura§ao	18	—	18	—	—
Djibouti	1	—	—	—	1
Dominica	1	—	—	1	—
Egypt	9	—	1	1	7
Eritrea	2	—	—	—	2
Fed. States of Micronesia	30	6	—	8	16
Fiji	51	1	15	3	32
France	3	—	—	1	2
French Polynesia	1	—	—	—	1
Greece	1	1	—	—	—
Guadeloupe	2	—	2	—	—
Guam	14	1	4	1	8
Honduras	7	3	1	—	3
India	19	—	14	2	3
Indonesia	82	5	5	8	64
Italy	5	1	—	2	2
Jamaica	7	—	1	2	4
Japan	113	7	46	13	47
Kiribati	1	—	1	—	—
Madagascar	3	1	1	—	1
Malaysia	3	—	—	—	3
Maldives	4	—	—	2	2
Marshall Islands	8	—	5	2	1
Mauritius	6	—	6	—	—
Mexico	32	2	14	—	16
Mozambique	2	—	—	—	2
New Caledonia	22	—	7	3	12
New Zealand	161	4	2	20	135
Nicaragua	1	1	—	—	—
Oceans	33	—	3	3	27
Palau	73	4	14	8	47
Panama	7	1	3	—	—
Papua New Guinea	152	2	60	3	87
Paracel Islands	5	1	1	—	3
Philippines	24	—	5	3	16
Puerto Rico	10	5	3	—	2
Russia	1	—	—	1	—
Seychelles	6	—	—	2	4
Singapore	4	—	1	—	3
Solomon Islands	14	—	2	1	11
Somalia	1	—	—	—	1
South Africa	13	—	2	1	10
South Georgia	1	—	—	—	1
South Korea	5	—	—	3	2
Spain	17	—	1	4	12
Taiwan	3	—	—	—	3
Thailand	18	—	4	1	13
Tonga	2	1	—	—	1
Trinidad & Tobago	2	—	—	—	2
Tunisia	2	—	—	2	—
USA	44	6	29	3	6
Vanuatu	3	1	—	—	2
Vietnam	1	—	—	—	1
Virgin Islands	2	—	—	1	1
Totals	1296	70	367	130	729
		5.4%	28.3%	10.0%	56.3%

Short-term intensive collections and long-term collections can influence the total number of collections for a country. For example, Ayling made biweekly collections of Aaptos aaptos at two sites in New Zealand for two and a half years (1976–1978).²⁷ Those 130 collections are greater than the total collections for all but two countries and more than twice the number of collections of the next most collected species. Similar examples are the 50 collections of Conus species in 1963 over two months in Thai and Indonesian waters and the 64 collections also of Conus species in 1997 from Papua New Guinea.²⁸ Not all of the collections were made to investigate MNPs; some collections were part of research projects on reef or species ecology or habitat extents.

Keeping all these caveats in mind, six countries had the most collections: New Zealand (161 collections; Ayling’s collections are 81% of the total), Papua New Guinea (152; Kohn’s collections are 42% of the total), Australia (147; 80% from the Great Barrier Reef), Japan (113), Indonesia (82), and Palau (73) (see also Figure SI#2–1).

Collections by Year.

The collection dates for the 1,296 collections span the years from 1945 to 2011, with the bulk occurring from 1976 to 2008 (Table 9). The temporal distribution of the collections, which peaked in the 1990s to early 2000s, is unsurprising given that our investigation focused on MNPs in more advanced stages of clinical testing. More recent collections tend to focus MNPs that are not included here. The potentially misleading effect of Ayling’s Aaptos aaptos collections (1976–1978)²⁷ can be seen if they are subtracted from the annual totals. Kohn’s 1963 and 1997 Conus collections²⁸ both create large peaks for those two years.

Table 9.

Collections by Year: Reported, Predicted, and Total (doubles omitted)

Year	Rep.a	Pred.b	Total	Year	Rep.a	Pred.b	Total	Year	Rep.a	Pred.b	Total
Pre-1960	2	0	2
1960	0	0	0	1980	8	1	9	2000	31	7	38
1961	0	0	0	1981	17	5	22	2001	13	5	18
1962	1	0	1	1982	9	4	13	2002	25	15	40
1963	50c	0	50	1983	15	5	20	2003	36	14	50
1964	0	0	0	1984	6	10	16	2004	50	2	52
1965	0	0	0	1985	24	9	33	2005	22	2	24
1966	0	0	0	1986	20	8	28	2006	35	4	39
1967	1	0	1	1987	20	18	38	2007	10	2	12
1968	2	0	2	1988	24	10	34	2008	32	0	32
1969	0	0	0	1989	21	12	33	2009	4	0	4
1970	1	0	1	1990	17	7	24	2010	4	0	4
1971	0	0	0	1991	36	15	51	2011	3	0	3
1972	2	1	3	1992	30	10	40	2012	0	0	0
1973	3	0	3	1993	32	10	42	2013	0	0	0
1974	2	2	4	1994	21	9	30	2014	0	0	0
1975	4	4	8	1995	28	14	42	Total	1035	261	1296
1976	39d	1	40	1996	27	20	47
1977	68d	2	70	1997	92c	11	103
1978	85d	2	87	1998	31	6	37
1979	13	11	24	1999	19	3	22

^acollection dates reported by source reference.

^bcollection dates predicted due to lack of reported collection date.

^cKohn collections of Conus spp.: 1963 – 50; 1997 – 64.

^dAyling collections of Aaptos aaptos: 1976 – 30; 1977 – 52; 1978 – 48.

Predicting Collection Dates.

As shown in Table 9 (bottom right corner), 261 collections (about 20 percent) were reported without collection dates. We explored the possibility of using the source article’s publication date as a predictor of the collection date by developing an ordinary least squares regression model. Despite the possibility that some of the reports are based on frozen specimens collected years prior to the publication, the model yields reasonable results (R² = 0.86) (details can be found in Section 3.2 of the Overview). The combined annual totals of collections with reported dates and with predicted dates are shown in Table 9.

Collections by Species.

We are using the term “species” in a broad sense to include specimens that could only be identified to the genus level. Of the 232 species reported here, 56 (24%) are only identified by their genus, and collectively they represent 166 (13%) collections. As described above and shown in Table 5, the dozen species with the greatest number of collections are Aaptos aaptos (137 collections), Moorea producens (64), Lamellodysidea herbacea (61), Luffariella variabilis (60), Theonella swinhoei (40), Rhabdastrella globostellata (35), Lissoclinumpatella (33), Dolabella auricularia (29), Ecteinascidia turbinata (25), Negombata magnifica (24), Mycale hentscheli (21), and Salinispora arenicola (20). Of the remaining species, 174 species (75%) have five or fewer collections, and 93 species (40%) have only one collection.

Collections by Phylum.

Most of the classification units we refer to as a phylum are in fact a phylum, but there are several exceptions that result from unsettled taxonomy or a disconnect between common associations and current taxonomic classifications: Fish is a paraphyletic group in the phylum Chordata;²⁹ Fungi is a kingdom within domain Eukaryota; Green Algae is a paraphyletic group of division Chlorophyta and division Charophyta; Rhodophyta is a phylum, but that classification appears to be an unsettled taxonomic issue;³⁰ and, Tunicata is a subphylum of phylum Chordata.³¹ Table 10 shows both the phylum totals and the percentage of total collections the phylum totals represent.

Table 10.

Collections by Phylum^a

Actinobacteriab	40	3.1%
Bryozoac	18	1.4%
Cnidariad	31	2.4%
Cyanobacteriae	77	5.9%
Dinoflagellataf	13	1.0%
Echinodermatag	12	0.9%
Fish	29	2.2%
Fungi	3	0.2%
Green Algae	8	0.6%
Hemichordatah	2	0.2%
Mollusca	195	15.1%
Nemerteai	3	0.2%
Poriferaj	716	55.2%
Rhodophytak	4	0.3%
Tunicatal	145	11.2%

^aFucoxanthin is found in most or all species of the classes Phaeophyceae (brown algae), Chrysophyceae (golden algae), and Bacillariophyceae (diatoms) in the phylum Ochrophyta plus some species in the phyla Dinoflagellata and Haptophyta, possibly as many as 16,000 species in total. These species and their collections are not included in the analyses presented here. See fucoxanthin entry in the MPT (SI#3) for references.

^bGram-positive bacteria.

^cmostly colonial filter feeders.

^dincludes octocorals.

^eex-blue-green algae.

^fmostly marine plankton.

^gincludes starfish, sea urchins, & sea cucumbers.

^hacorn worms.

ⁱribbon worms.

^jsponges.

^kred algae.

^ltunicates.

Collections by Clinical Status.

As described above, we categorized MNPs according to the most advanced stage of pharmacological investigation they reached for any of their potential therapeutic uses. For collections of specimens that yielded multiple MNPs having different clinical status, we assigned them to the most advanced category. For example, collections of Aplysinella rhax, the source of panobinostat (Approved) and psammaplin A (Lead/Probe), are categorized as Approved. The distribution of collections can be found at the bottom of Table 8, which also provides the clinical status distributions of collections by country.

COMMON ERRORS AND OMISSIONS FOUND IN COLLECTION REPORTS

As has been described above, collections data reported by source references sometimes contain errors of commission or omission. We attempted to correct errors and surmise missing data. In some instances, this meant including two or more possible locations for a collection; in others, the number of collections or the collection location is just our best estimate. We created some broad categories and subcategories (Table 6) to organize prominent examples, which can be found in Table SI#1–3, which lists over 165 such examples. Some of the problems are unavoidable, such as when a specimen’s taxonomic classification is revised or the collection location’s placename is changed. Other problems are created by advances in technology, such as the significant advances in geolocation technology that make pinpoint determinations of location possible with a handheld device or the click of a mouse. A goodly number of problems (mainly omitted data) likely result from a belief that the data are unimportant or unnecessary. Finally, there are a small number of typographic errors that were missed by the many pairs of eyes that read the text prior to publication (such as arcminute values greater than 60).

It is understandable that the focus of most investigators is on issues related to the chemistry and pharmacology of the MNPs, but we would urge them to strive for complete and accurate reporting of their collections (e.g., use BGN Approved Names or Recognized Name Variants, report accurate lat-lon coordinates, report dates of collections, etc.). Blunt and colleagues and Leal and colleagues have made similar suggestions in the past.³²

DEVELOPING A METRIC OR MODEL OF PHARMACOLOGICAL POTENTIAL

The large quantity of data and information in this series of documents makes summarization difficult, but we would suggest that several observations can be made:

• MNPs currently provide ecosystem services as clinically useful drugs and probes, and potentially many more MNPs could become pharmacologically useful.

• Consequently, benefits assessments of marine ecosystems, particularly coral reefs, should include existing and potential pharmacological benefits from MNPs.

• MNPs with pharmacological potential have been found in marine ecosystems of every continent and every ocean.

• MNPs with pharmacological potential are found in many marine phyla, but Porifera (sponges) and microorganisms (including symbiotic microorganisms of Porifera and other phyla) are the most frequently identified sources for the collections reviewed here.

The data reported here represent the first, necessary step in a multi-step process of exploration. The data lay the foundation for the development of methods for predicting, qualitatively or quantitatively, a location’s pharmacological potential and the ecosystem services that could arise therefrom. The data may afford the possibility of defining pharmacological potential in terms of either a single endpoint (FDA approval) or a combination of mid-points and endpoints (e.g., an experimentally useful lead or probe, clinical testing but not approval, etc.). One avenue of exploration would combine the collection location data with the species distributions to pinpoint hotspots within the species distribution and to identify inconsistencies that merit further examination. This path could reduce uncertainties related to taxonomic and specimen identification. Another avenue of investigation would combine the spatial collections data with reef condition data and predictions of reef condition to explore whether there is a relationship between reef condition and pharmacological potential. If such a relationship is discovered, it could lead to a metric or model for predicting a location’s pharmacological potential.

LONG-TERM MAINTENANCE OF THE DATA

Large data sets such as the one described in this article are ephemeral unless they are maintained and regularly updated. If the data contained in the Supporting Information documents are seen as sufficiently valuable to warrant regular updating, it will be incumbent on the MNP community to devise a means of accomplishing that (perhaps as Faulkner and then Blunt and colleagues provide annual surveys of new MNPs).