Genetic Sexing of the African Penguin, Spheniscus demersus Using Noninvasive Guano and Molted Feather Samples

TL;DR

Researchers developed a noninvasive way to determine the sex of African penguins using guano and molted feathers, aiding conservation efforts.

Contribution

A noninvasive DNA extraction and PCR-based sexing method for African penguins using guano and molted feathers was optimized.

Findings

Two primer sets targeting CHD1-Z and CHD1-W genes were successfully used for genetic sexing.

Optimized methods enabled DNA extraction from inhibitor-rich guano and difficult-to-lyse molted feathers.

The method supports conservation monitoring of wild African penguin populations without invasive sampling.

Abstract

The monitoring of sex ratios in wild populations of the critically endangered African penguin
Spheniscus demersus
is essential for conservation management but is currently limited by the inherent difficulty in acquiring blood samples required for sexing. This study optimized a noninvasive method for the DNA extraction and PCR‐based genetic sexing of
S. demersus
using guano and molted feather samples. Two primer sets (CHD1F/R & 2550F/2718R) were used that target sex‐specific length polymorphisms in the CHD1‐Z and CHD1‐W genes on the CHD1 sex chromosomes. Using methods optimized for the extraction of DNA from inhibitor‐rich guano and difficult‐to‐lyse molted feather samples, this work can directly contribute to the conservation monitoring of wild
S. demersus
populations through sex determination using noninvasive means.

This study developed a noninvasive method for the DNA extraction and PCR‐based genetic sexing of
S. demersus
using guano and molted feather samples. Two primer sets (CHD1F/R & 2550F/2718R) were used that target sex‐specific length polymorphisms in the CHD1‐Z and CHD1‐W genes on the CHD1 sex chromosomes. Using methods optimized for the extraction of DNA from inhibitor‐rich guano and difficult‐to‐lyse molted feather samples, this work can directly contribute to the conservation monitoring of wild
S. demersus
populations through sex determination using noninvasive means.

Figures

Gel electrophoresis banding patterns for the CHD1‐Z and CHD1‐W genes amplified from DNA in Spheniscus demersus blood, guano, and feather samples. F = female; M = male. Ladder 50 bp (Zymo Research; Cat. No. M5004‐50).

Tables

TABLE 1. Two primer sets used for the genetic sexing of Spheniscus demersus using DNA extracted from blood, guano, and feather samples.

Primer	Sequence (5′‐3′)	References
CHD1 F	TATCGTCAGTTTCCTTTTCAGGT	Lee et al. (2010)
CHD1 R	CCTTTTATTGATCCATCAAGCCT
2550 F	GTTACTGATTCGTCTACGAGA	Fridolfsson and Ellegren (1999)
2718 R	ATTGAAATGATCCAGTGCTTG

Full text

Introduction

1

Wild populations of the African penguin Spheniscus demersus have experienced a near‐constant decline since their vulnerable listing on the International Union for Conservation of Nature (IUCN 2024) Red List in 1984. As of 2024, the species has been uplisted to a critically endangered status, with an estimated number of breeding pairs falling below 10,000, inviting the possibility that S. demersus may succumb to extinction in just over 10 years (Sherley et al. 2024). Population decline has been linked to environmental and anthropogenic stressors, such as habitat destruction (Frost et al. 1976; Sherley et al. 2014a), overfishing of prey (Crawford et al. 2011), and parasitic infection (Espinaze et al. 2019); all of which are compounded by the complicated stressors of a changing climate (Weller et al. 2014; Dreyer et al. 2024).

To aid in the conservation management of this species, further steps must be taken to increase population monitoring in the wild, including the accurate and efficient quantification of sex ratios. Population sex ratios are crucial for understanding wild population dynamics as they influence breeding rates, parental care roles, genetic diversity, and overall population stability and recovery (Rosa et al. 2017; Liker et al. 2013). Analyses of adult sex ratios in wild bird populations show many species are male‐biased (Payevsky 2021), which has been linked to competition effects and a reduction in offspring that can threaten population viability (López‐Sepulcre et al. 2009). In S. demersus , a previous study on sex determination via necropsy showed that mortality among juvenile and adult birds is largely female‐skewed (> 60%), which may lead to a male‐biased sex ratio and further contribute to the rapid rate of decline for the species (Pichegru et al. 2013). Understanding sex ratios of wild populations could contribute to conservation practices by tailoring effort and resources to the needs of more sex‐imbalanced colonies (e.g., chick‐bolstering, artificial nest installation, habitat restoration). These data would also be vital to colony establishment projects (Smith 2021) as well as for the selection of rehabilitation release sites, to encourage breeding pairs in wild populations.

However, sex determination in wild S. demersus is difficult, as the species exhibits only slight external sexual dimorphism in beak and body size that remains an inconsistent indicator of sex due to the overlapping morphometric ranges among females and males (Campbell 2015). Further, steroid sexing methods using estrogen to testosterone ratios are unreliable for animals prior to sexual maturation and during breeding periods (Bautista et al. 2013) and have not yet been validated for S. demersus . Thus, sex determination in live S. demersus is currently established via avian laparoscopy and cloacal examination, or through the genetic testing of invasive blood or tissue samples, which are inherently difficult to obtain from wild animals and can cause stress and possible injury (Parsons et al. 2015).

While genetic sex determination from bird feces has been accomplished for many species (Asawakarn et al. 2018; Segelbacher and Steinbruck 2001), including from some penguin species (Gentoo penguin, Tian et al. 2021), genetic sex determination in guano samples from S. demersus has not yet been published. Similarly, genetic sexing from feathers has been performed for other bird species (Guaricci et al. 2025; Bello et al. 2001; Sacchi et al. 2004; Morandini et al. 2024; Mori et al. 2020), including a Spheniscus congener (Humboldt penguin; Constantini et al. 2008), but these methods have not yet been confirmed for S. demersus . Further, these previous studies used plucked or otherwise physically removed feathers, which has been argued to be a discouraged practice in avian literature and comparable to the ethical implications of blood sampling (McDonald and Griffith 2011). There are also a few existing publications (Dawson et al. 2015; Griffiths and Tiwari 1995) that focus specifically on genetic sexing using molted or otherwise shed feathers.

In the following study, PCR‐based genetic tests validated the use of noninvasive guano and molted feather samples to accurately determine sex in S. demersus individuals under human care. Sex identification was achieved using an optimized method for DNA extraction from inhibitor‐rich guano and archival molted feather samples, combined with the targeting of two primer sets (CHD1F/R & 2550F/2718R; Table 1). Both primers were effective in the visualization of sex‐specific intronic variations reflected by variable length polymorphisms between CHD1‐Z and CHD1‐W regions of the chromobox‐helicase DNA‐binding (CHD) sex chromosome. Further, these methods were used for the reliable sex determination of two S. demersus chicks (≤ 2 months old) via the collection of their first guano sample, indicating the utility of these methods in the sexing of animals prior to their sexual maturation, when other sexing methods are unreliable or unusable. This work can contribute to increased sex ratio monitoring of S. demersus populations using noninvasive samples that will ultimately inform and aid in the management of their conservation in the wild.

TABLE 1: Two primer sets used for the genetic sexing of Spheniscus demersus using DNA extracted from blood, guano, and feather samples.

Methods

2

Sample Collection

2.1

Guano and molted feather samples were collected opportunistically from Spheniscus demersus individuals under professional human care at Mystic Aquarium, Mystic, CT. In addition, blood samples used as positive controls in DNA extraction and PCR were collected from the same animals during veterinary exams. The tested guano and feather samples were collected from individual birds between 2021–2024 and 2020–2024, respectively, while all blood samples were collected in 2024. Individual birds were identified by an external marker band system affixed to the right wing. Guano samples were collected from two chicks between 5–8 weeks old, while guano, molted feather, and blood samples were collected from 12 adult birds between 10–37 years old. All S. demersus chicks sampled were reared from breeding pairs at Mystic Aquarium as a part of the Association of Zoos and Aquariums Species Survival Plan and are under permanent human care (Sarro and Sirpenski 2020).

Guano samples were collected opportunistically directly from the African Penguin habitat. An individual bird was observed until defecation occurred, at which point the sample was collected using a Puritan HydraFlock Sterile Flocked Swab (Fisher Scientific; Cat. 22‐029‐625) and the animal ID and date were logged. Samples that were visibly apparent to have a higher ratio of feces (i.e., solid, green) to urate (i.e., liquid, yellow) content were preferentially collected when possible to avoid PCR inhibitors present in high‐concentration urate samples (Segelbacher and Steinbruck 2001). Duplicate swabs were collected for each fecal sample and were immediately placed on ice until archival storage at −80°C.

All molted feathers were collected opportunistically and noninvasively from birds during their respective molting period. Feathers were collected when shed from individual birds undergoing molt, when they were isolated prior to routine veterinary exams. Feathers were stored in plastic Ziploc bags with date and animal ID and archived at room temperature in a dry dark drawer until their extraction in 2024. Two feathers were collected from each animal tested.

Approximately 3 mL of blood was collected into an Ethylene Diamine Tetra Acetic acid (EDTA) coated vacutainer and preserved at −20°C. Alternatively, if DNA extraction was anticipated within 12 h of sample collection, the sample was kept at 4°C.

DNA Extraction

2.2

All methods of DNA extraction were carried out in an area free of post‐PCR products. Guano (from 14 birds) and molted feather (from 12 birds) samples were run in duplicate, while a single blood sample was run for each adult individual tested (12 total) as a positive DNA extraction control. All 12 adult birds had paired guano, feather, and blood samples. Only guano samples were retrieved from the two chicks, with blood samples taken later by veterinary staff to be confirmed by clinical means. Negative controls of ddH2O template were used for each primer set.

Archived guano swab samples were thawed from −80°C storage at room temperature, and the 3–3.5 cm swab tip was cut using 10% bleach‐sterilized scissors and placed into a sterile 1.5 mL tube, which was then spun down briefly at 8000 RPM for 1 min. Any visible liquid urine supernatant was removed and discarded. Approximately 10–60 mg (the lower range of this was used for samples that appeared to have a high urate concentration) of remaining solid fecal sample was then transferred to a sterile 1.5 mL tube, and 200 μL of lysis buffer (10 mM Tris–HCL pH 7.0; 2 mM EDTA; 1.5% SDS), and 15 μL of Proteinase K were added and vortexed briefly before a 24‐h incubation in a 55°C continually‐agitated water bath, with an addition of 5 μL Proteinase K and a 1 min vortex of each sample halfway between incubation (i.e., after approximately 12 h). Immediately following this pretreatment, the QIAamp DNA Stool Mini kit (Qiagen AB; Cat. No. 51604) was used for DNA extraction following manufacturer's instructions for human DNA analysis, except for a halving of all volumes other than the InhibitEx reagent (i.e., this volume was not reduced) and a reduced final elution volume of 50 μL.

For molted feather samples, 10% bleach‐sterilized scissors and tweezers were used to separate the entire calamus portion of a single molted feather (~20 mm) and trim lengthwise and then widthwise into as small fragments as possible, and placed into a 1.5 mL sterile tube. Next, 500 μL of lysis buffer (1 M pH 8.0 Tris–HCl, 25% SDS, 25 mM EDTA), 40 μL of Proteinase K, and 30 μL of 1 M DTT (dithiothreitol) were added to the tube and incubated for 24–48 h in a 55°C continually agitated water bath. In most samples, after 24 h, there was still undigested material visible, in which case an additional 10 μL of Proteinase K and 5 μL of DTT was added, vortexed for 1 min, and then incubated for 12–24 more hours. Following incubation, samples were brought to room temperature and then immediately extracted using the DNeasy Blood & Tissue Kit (Qiagen; Cat. No. 69504), except for a reduced elution volume of 50 μL. Blood samples were brought to room temperature, and an 8 μL aliquot was transferred to a 1.5 mL sterile tube. DNA extraction was performed using the DNeasy Blood & Tissue Kit (Qiagen; Cat. No. 69504), except for an extended incubation time of 3 h in a 56°C water bath, and a reduced final elution volume of 100 μL.

PCR Amplification

2.3

PCR was performed with a 24 μL final volume for each reaction and contained 12.5 μL (10X) DreamTaq PCR Master Mix (Thermo Scientific; Cat. No. K1072), 1 μL of each forward and reverse 10 μM primer, 2 μL of DNA template, and 7.5 μL ddH2O. For both primer sets, PCR amplification included an initial pre‐denaturation of 95°C for 5 min, 10 cycles of 94°C for 30 s, 56.8°C to 48.7°C (at a gradient step‐down decrement of 0.9°C per cycle), and 72°C for 30 s, and then 34 cycles of 94°C for 30 s, 48°C for 30 s, and 72°C for 30 s, with a final extension of 72°C for 5 min. Visualization of PCR product was completed via gel electrophoresis in a 2.5% gel using 1X Tris Borate EDTA (TBE) buffer at 80 V for 60 min. Individual bands of positive amplification products were isolated from the gel and purified using Zymoclean Gel DNA Recovery Kit (ZymoResearch; Cat. No. D4001) and sent for Sanger sequencing via EuroFins Scientific.

Results

3

While both guano and blood samples had 100% success in amplification, only about 87%–88% of molted feather samples showed PCR products (three samples showed no bands) with varying levels of gel band brightness. Amplification failure of molted feather samples was not associated with the age of archived samples, as the oldest samples tested (~4 years old) showed positive bands.

Female S. demersus were identified using gel electrophoresis by two bands at 515 (CHD1‐Z) and 600 bp (CHD1‐W), while males were identified by a single 515 bp (CHD1‐Z) band (Figure 1) using CHD1 primers (Lee et al. 2010). The 2550F/2718R primers (Fridolfsson and Ellegren 1999) also showed a visible difference between female and male samples, where females were observed to have two bands, at 615 bp and 700 bp, while males only had a single band at 415 bp (Table 1).

Gel electrophoresis banding patterns for the CHD1‐Z and CHD1‐W genes amplified from DNA in Spheniscus demersus blood, guano, and feather samples. F = female; M = male. Ladder 50 bp (Zymo Research; Cat. No. M5004‐50).

Guano samples taken from two chicks were used to accurately determine the sex of both individuals, prior to confirmation of sex with clinical blood tests taken a month later. The closest sequence results from amplification of CHD1 Z and W showed a 97% and 83% identity match with sequences from the CHD1 Z and W genes, respectively, of the Magellanic penguin ( Spheniscus magellanicus ) on GenBank (Accession: CHD1Z: GU451227; CHD1W: GU451231).

Discussion

4

Increased demographic and ecological monitoring of wild S. demersus populations is essential in the conservation management of this critically endangered species, as only limited data exist (Pichegru et al. 2013; Spelt and Pichegru 2017). Sex determination provides useful data in the estimation of reproductive stability and conservation management of endangered species (Çakmak et al. 2017; Donald 2007), yet sex discrimination techniques remain inherently difficult or unreliable. The use of noninvasive guano and molted feather samples in the PCR‐based genetic sexing tests validated in this study presents promising utility for the collection of these data from wild populations without the use of invasive blood tests, which currently limits the number of individuals used to estimate sex ratios.

A primary obstacle in this study was the extraction of DNA from guano and molted feather samples. Unlike pure fecal samples from most animals, guano also contains urate, a known PCR inhibitor (Segelbacher and Steinbruck 2001) that can suppress PCR target amplification. The extraction method here is specifically modified to reduce PCR inhibition from guano samples, including an extended incubation period and use of a commercial extraction kit designed specifically for PCR inhibitor depletion.

Molted feather samples proved similarly difficult to extract DNA from, as evidenced in previous studies (Segelbacher 2002; Griffiths and Tiwari 1995; Dawson et al. 2015) as they have been shown to contain less viable and lower volumes of total DNA than those of the calamus of freshly plucked or otherwise physically removed feathers. Further, DNA often degrades over time, and any DNA present is difficult to lyse as it is generally contained in keratin sheathing. The use of DTT and an extended digestion and lysing period proved essential for sufficient DNA extraction.

Although useful for birds of all ages, the sensitivity and accuracy of the guano sexing test methods may be of particular importance for S. demersus chicks. While guano samples tested in this study visibly appeared to have higher concentrations of urate, all samples still provided accurate confirmation of sex. A major issue in the sexing of S. demersus chicks (including those reared under professional human care) using current methods is that it can take months before the animal can safely be restrained to take a blood sample for sex determination. Guano samples can instead be collected opportunistically and noninvasively, and thus can be analyzed as soon as the first defecation upon hatching. There are inherent issues with the collection of DNA in guano samples from S. demersus chicks directly from nesting areas in the wild, including the cross‐contamination of parent guano DNA and the higher ratio of urate to feces present in chick guano, the latter making the removal of urate in a sample more difficult than that from adults. However, as the practice of hand‐rearing chicks in wild rehabilitation centers increases (Sherley et al. 2014b), these methods may prove useful in collecting sex data prior to release that would otherwise be challenging to collect via blood sample, and may aid in elucidating the projection of future sex ratios.

The ability to use molted feathers for sexing may be of particular value for archived time series samples, as samples up to 4 years old were able to yield enough DNA to amplify and visualize via gel electrophoresis (samples > 4 years old were not tested). For rehabilitation centers that work with wild S. demersus , these samples are also easy to store, as the samples used in the current study were not preserved in any particular means other than being kept in a drawer and protected from light at room temperature. However, there is an inherent issue in using molted feathers to study individuals among truly wild colonies, as collecting molted feathers in the environment and assigning them to a single individual is logistically impractical. One way to utilize this method appropriately may be through the collection of feathers from molting S. demersus individuals that are regularly admitted to rehabilitation centers (e.g., Sherley et al. 2014b), where molting adults may be isolated for periods of time.

Overall, the collection of guano or molted feathers for the genetic sexing of wild S. demersus individuals offers similar issues, as both can be difficult to collect and assign to a single animal without contamination of surrounding birds. However, genetic sexing with guano samples appears to be a more reliable and more logistically feasible noninvasive sample source than molted feathers at present.

Ultimately, the PCR‐based genetic sexing validated in this study presents the first accurate and reliable technique for sex determination of S. demersus using entirely noninvasive means. The use of these methods offers a practicable utility for the demographic monitoring of wild S. demersus individuals and thus contributes to informing the conservation of this now critically endangered species.

Author Contributions

Susan A. Smith: conceptualization (supporting), data curation (lead), investigation (lead), methodology (lead), writing – original draft (lead), writing – review and editing (equal). Maureen V. Driscoll: conceptualization (lead), funding acquisition (lead), writing – review and editing (equal). Tracy A. Romano: conceptualization (equal), funding acquisition (lead), project administration (lead), resources (equal), supervision (equal), writing – review and editing (equal).

Funding

This work was supported by the Association of Zoos and Aquariums, CGF21‐1744.

Conflicts of Interest

The authors declare no conflicts of interest.