Development and characterization of microsatellite markers in the small Indian mongoose (Urva auropunctata)

The small Indian mongoose (Urva auropunctata) is one of the world’s worst invasive alien species and eradication programs are ongoing worldwide. The development of individual and sex identification markers will improve their management. We searched for novel mongoose microsatellite markers using genome-wide screening and identified 115,265 tetra-nucleotide repeat loci. Of 96 loci tested, 17 were genotyped in 28 mongooses from the Okinawa population. The genetic diversity analysis showed that the average expected and observed heterozygosity and number of alleles were 0.55, 0.56, and 2.94, respectively. Of 17 loci, one deviated from Hardy–Weinberg equilibrium and six loci pairs were likely linked to each other. However, we succeed in identifying all individuals using all of the microsatellite loci. The novel sex identification markers worked successfully in a test using sex known samples. Our novel microsatellite and sex identification markers should be useful in studies of individual identification and population genetics of the mongoose.


Introduction
The small Indian mongoose Urva auropunctata (Hodgson, 1836) is a generalist predator that naturally occurs from the Arabian Peninsula to Southeast Asia [1]. It was introduced to 64 islands in the Pacific and Indian Oceans and Caribbean and Adriatic Seas and to mainland Europe, South America, Australia, and North America for rat or snake control, causing enormous damage to the native biodiversity [2]. Ecological niche modelling of the mongoose suggests that it will spread globally by 2050 [3].
Mongoose eradication programs are underway using trapping and poisoning. On Amami Oshima Island, Japan, the program is close to success [4], and the populations and distributions of the endemic Amami rabbit Pentalagus furnessi (Stone, 1900) and Amami spiny rat Tokudaia osimensis (Abe, 1933) have expanded [5,6]. To continue these eradication programs, methods to estimate the mongoose population are needed.
The application of genetic analysis is essential for wildlife management. Microsatellite markers are often used to reveal the genetic diversity and origin of invasive species through population genetic analysis [e.g., 7,8]. Single nucleotide polymorphisms (SNPs) are also powerful markers in population genetics; however, SNP analysis requires many markers to be analysed than microsatellite analysis [9], and it is unsuitable for analysing additional samples. Therefore, microsatellite markers were chosen in the present study.
Previously, eight mongoose microsatellite markers were developed [10]; however, there is a problem with their use. Three of these loci could not be scored reliably [11], implying that the loci included unstable repeat characters. This problem leads to mis-genotyping and reduces the resolution power of markers.
This study developed novel microsatellite markers for the small Indian mongoose using genome-wide screening in a mongoose from the Okinawa, Japan, population. We selected tetra-nucleotide repeats with short amplification lengths as markers. We also developed sex identification markers. The novel markers were verified with 28 mongooses from the Okinawa population.

Sample collection
We obtained tissue samples from 28 mongooses in the Okinawa population ( Fig. 1; Table 1). All samples were taken from euthanized animals through the mongoose eradication program.

Genome sequences and identification of microsatellite loci
Genomic DNA was extracted from a small Indian mongoose tissue sample (Nago1; Table 1) and the genome was sequenced by Macrogen Japan (Kyoto, Japan). Wholegenome shotgun sequencing was performed on an Illumina NovaSeq 6000 sequencer (150 paired ends). The raw sequence data was deposited in the DNA Data Bank of Japan (DDBJ) Sequence Read Archive (DRA; https:// www. ddbj. nig. ac. jp/ dra/) with the accession number DRR294934. The sequence quality was checked by the FastQC program, and noise and adapters were trimmed with fastp [12] under the default configurations. We then mapped the sequences onto the genome of a banded mongoose Mungos mungo (acc. no. GCA_004023785) using BWA-mem2 [13] with the default parameters. Among the species with genome information, M. mungo is most closely related to the small Indian mongoose [14]. To produce the consensus sequence, the output BAM file was sorted using SAMtools [15] and the variants were called with bcftools. In total, 52,502 sequences were obtained, with a total valid sequence length of 1,435,707,730 bp.
The microsatellite loci were identified from the consensus sequences using Krait [16] with the following configuration: perfect microsatellite search mode; minimum repeat number of tetra-nucleotide repeats = 7; and flanking sequence length = 200. We selected tetra-nucleotide repeats because the repeats are easy to determine and the genotype is less prone to slippage [17]. To search for primers, Primer3 [18] implemented in Krait was used with following The other parameters were set to the default values. In total, 115,265 tetra-nucleotide repeat loci were found and 10,224 primer pairs were exposed. Of these, 96 primer pairs were selected for multiplex PCR amplification based on their product sizes, which were around 80, 120, 160, or 200 bp. To avoid linkage disequilibrium, a marker was selected from each scaffold.

Genotyping microsatellite loci
To verify the novel microsatellite markers, 28 tissue samples were used (

Development of sex identification markers
Based on the sequences of zinc finger protein genes on the X (ZFX; AB848712) and Y (ZFY; AB848711) chromosomes from Murata et al. [21], new mongoose sex identification markers were designed (

Molecular data analyses
For all microsatellite loci, the observed (Ho) and expected (He) heterozygosity, deviation from Hardy-Weinberg equilibrium (HWE), inbreeding coefficient (Fis), and linkage disequilibrium were analysed using GENEPOP [22]. The presence of null and dropout alleles was tested with Microchecker [23]. The probabilities of identity (PID and PIDsib) were calculated by GIMLET [24].

Results and discussion
Of 96 loci tested, 18 produced single bands of the target size. All loci except one successfully scored the genotypes by fragment analysis, and two to five alleles were found in the Okinawa mongoose population ( Table 2). The 17   (Table 2). Microchecker suggested that no loci showed evidence of null alleles, while one locus (Uau16) deviated from HWE expectations (P < 0.003). Based on linkage disequilibrium tests, six loci pairs were likely linked (P < 0.05): Uau1 and 17, Uau1 and 7, Uau1 and 8, Uau3 and 7, Uau6 and 13, and Uau7 and 10. Generally, the genetic diversity of invasive species is lower than in the native population owing to the small number of founders or bottlenecks. There were at most 17 founders of the Okinawa mongoose population [25]. This history suggests that the Okinawa population might not follow HWE expectations. Using all 17 loci, 28 individuals were identified genetically; the PID was 1.92 × 10 −10 and the PIDsib 3.68 × 10 −5 . At minimum, four loci were required to reach PID < 0.001 and seven loci were required to reach PIDsib < 0.01. This suggests that our novel microsatellite markers are useful for individual identification.
We also confirmed that the novel sex identification markers worked successfully in a test using known sex samples ( Figure S1).
As a result, we successfully developed 17 microsatellite markers and sex identification markers. These markers should be powerful tools for managing mongooses in combination with DNA collected from non-invasive samples, though it needs to verify and optimize the protocols.