72 
Fishery Bulletin 112(1) 
and this has ultimately resulted in limited resolution 
in addressing population structure and hybridization 
among Brevoortia congeners in the Gulf of Mexico (An- 
derson, 2007; Anderson and Karel, 2007; Anderson and 
McDonald, 2007). 
The first step in addressing population structure 
among wild populations is the identification of appro- 
priate polymorphic loci. Adequate loci with sufficient 
polymorphism are required to definitively detect pat- 
terns of divergence from limited sample sizes. When 
low levels of genetic divergence have occurred between 
populations, as might be expected in Gulf Menhaden, 
the total number of gene loci employed, as well as the 
average number of alleles across loci, is predictive of 
the power in detection of that divergence (Kalinowski, 
2004; Bernatchez and Duchesne, 2000). 
Additionally, gene loci should have a limited amount 
of extraneous “noise” that can be created by assay- 
specific genotype errors (for instance, size homoplasy, 
null alleles, or allele dropout). Many published primer 
sets intended for use with wild populations generally 
consist of an initial data analysis from a single, small 
sample (n<20 individuals) and, therefore, may either 
lack polymorphism or include one or more genotyp- 
ing inconsistencies that are not identified in the ini- 
tial marker characterization. Furthermore, primer sets 
for nonmodel organisms can be difficult to find in the 
published literature, and the lack of such primer sets 
commonly results in use of heterologous gene loci from 
model organisms or ecologically important congeners 
(Rico et al., 1996). Heterologous loci have been used 
commonly in nonmodeled organisms, but the genetic 
variability of these loci in nontarget species is usually 
lower than the genetic variability in target species (El- 
legren et al., 1997) and the evolutionary properties of 
the repeat motif itself (such as motif sequence or pat- 
tern) are unknown. 
This study attempts to address the lack of adequate 
genetic loci for the genus Brevoortia generally and 
for Gulf Menhaden in particular. A genomic library of 
Gulf Menhaden restriction fragments was enriched for 
novel short tandem repeats (STRs, or microsatellites), 
and microsatellite markers were identified by using a 
published method (Glenn and Schable, 2005). The dis- 
criminatory power of the resulting group of microsatel- 
lite loci in the Gulf Menhaden population assignment 
was quantitatively assessed, as was the likelihood of 
species identification through the use of a subset of di- 
agnostic loci. These loci increase the genetic tools avail- 
able for improvement of stock definitions in the Gulf 
Menhaden and other species of Brevoortia. 
Materials and methods 
Study area and sampling methods 
Specimens of Brevoortia spp. were collected through- 
out the North American range of this genus and were 
identified using the morphological characteristics de- 
scribed in Dahlberg (1970). For the initial phase of the 
study (microsatellite identification), genomic DNA was 
isolated from 2 Gulf menhaden juveniles (<40 mm to- 
tal length) collected inshore in the vicinity of Freeport, 
Texas (Fig. 1) in spring 2012. These individuals were 
collected with a bag seine along the shoreline. The sec- 
ond phase of the study was carried out with previously 
sampled individuals from Anderson (2007). These fish 
included 118 Gulf Menhaden individuals from 4 sam- 
pling locales in the Gulf of Mexico sampled between 
2002 and 2003: Laguna Madre, Texas, 2003 [n=29]; 
Matagorda Bay, Texas, 2003 [re=30]; Lake Charles, 
Louisiana, 2002 [n= 29]; and Apalachicola, Florida, 2003 
[ft =30]). Samples also included 30 Atlantic Menhaden 
(Brevoortia tyrannus) from Sandy Hook Bay, New Jer- 
sey, collected in 2002; 8 Atlantic Mehaden from Bogue 
Sound, North Carolina, collected in 2002; 22 Finescale 
Menhaden from Sabine Pass, Texas, collected in 2004; 
and 19 Yellowfin Menhaden from Charlotte Harbor, 
Florida, collected in 2003. Because these samples were 
collected across multiple years, it was expected that 
multiple independent cohorts were collected. 
Library enrichment and microsatellite loci cloning 
A microsatellite-enriched library of Gulf Menhaden 
genomic DNA was constructed and microsatellite loci 
were cloned by using the protocol of Glenn and Schable 
(2005). Genomic DNA was isolated from 2 juvenile Gulf 
Menhaden sampled in Freeport, Texas, with a Gentra 
Puregene Tissue Kit (Qiagen, Inc., Valencia, CA 2 ), and 
DNA was eluted in 50 pL of TE buffer (10 mM Trig, 1 
mM EDTA). Each DNA sample was then cut with Rsal 
restriction enzyme, resulting in fragments in a range of 
200-1000 base pairs (bp). Restriction fragments from 
each sample were then combined into a single pool and 
ligated to biotin-labeled SuperSNX linkers (Glenn and 
Schable, 2005). Linked fragments were enriched for mi- 
crosatellite repeat-bearing fragments with biotin-cap- 
ture Dynabeads (Life Technologies Corp., Carlsbad, CA) 
in 2 enrichment mixtures: a trinucleotide mix (AAC, 
AAG, AAT, ACT, ATC) and a tetranucleotide mix num- 
ber (AAAC, AAAG, AATC, AATG, ACAG, ACCT, ACTC, 
ACTG). Successive temperature steps of T m i=48.5°C, 
and T m 2=58.5°C were used for enrichment. Enriched 
fragments were ligated into plasmids and transformed 
by using Invitrogen TOPO-TA cloning kits (Life Tech- 
nologies Corp.), according to the manufacturer’s in- 
structions. Transformed bacteria were plated onto 150- 
mm ampicillin-inoculated S-Gal agar plates (Sigma-AI- 
drich Co., St. Louis, MO) and grown for 2 days at 37°C. 
Positive colonies were selected and stored in 50 pL of 
TE buffer before DNA sequencing. 
2 Mention of trade names or commercial companies is for iden- 
tification purposes only and does not imply endorsement by 
the National Marine Fisheries Service, NOAA 
