Evaluation of dimethylformamide (DMF) and Trehalose as cryoprotectants in African penguin Spheniscus demersus semen cryopreservation

Abstract

We investigated the use of dimethylformamide (DMF), a permeating cryoprotectant, and its combination with trehalose, a non-permeating cryoprotectant, for the cryopreservation of African penguin (Spheniscus demersus) semen. Semen samples were cryopreserved using either 6% DMF alone (control group) or 3% DMF with 500 mM trehalose (T500 group). Post-thaw sperm quality was assessed based on motility parameters, viability, mitochondrial membrane potential, acrosomal integrity, apoptosis, and DNA fragmentation. Each treatment exhibited significant differences in specific aspects of post-thaw sperm quality. The control group showed higher total motility and mitochondrial membrane potential, whereas the T500 group demonstrated improved kinematic parameters (VCL, VSL, VAP, ALH, BCF), reduced apoptotic-like changes, less acrosomal damage and lower DNA fragmentation. Both treatments maintained viability, with a strong correlation between morphological and fluorescent viability assays. While the addition of trehalose did not enhance total motility, it conferred targeted cellular protection and improved motion dynamics potentially relevant to fertilization. These findings indicate that DMF and trehalose can improve post-thaw semen quality in African penguins, and that cryoprotectants may influence specific sperm traits differently across and within species.

Introduction

Among the 18 recognized penguin species, the African penguin (Spheniscus demersus) is one of the most endangered^1,2. In 2024, its IUCN status was uplisted from Endangered (EN) to Critically Endangered (CR) due to a rapid and ongoing population decline^3,4. As the only penguin species native to South Africa, it has faced decades of anthropogenic pressure^5–7. Guano harvesting severely degraded its nesting grounds, while extensive egg collection further contributed to population loss^6,7. Today, oil spills, maritime traffic noise, and commercial overfishing remain major threats, accelerating the species’ decline^2,7–10.

Numerous institutions are actively engaged in in situ conservation efforts focused on the rescue, rehabilitation, and release of injured, oiled, toxicated, or malnourished adults and chicks^2,7,10–12. However, given the projected extinction of the species in the wild by 2035^13–15, integrating genetic biobanking into conservation programs is essential to ensure long-term genetic diversity^16–20.

Only a few studies have attempted to develop species-specific sperm cryopreservation protocols for penguins, and even fewer have focused on African penguins, each employing different cryoprotective agents^21–26. In Magellanic penguins (Spheniscus magellanicus), dimethyl sulfoxide (DMSO) and ethylene glycol (EG) were used²¹. In Gentoo penguins (Pygoscelis papua), glycerol, dimethylacetamide (DMA), and DMSO^22,24 were tested, while in King penguins (Aptenodytes patagonicus), DMSO served as a cryoprotectant²⁵. For African penguins, cryopreservation studies have explored glycerol²², DMA and DMSO²⁴.

Glycerol, DMA, DMSO, and EG are permeating cryoprotective agents (P-CPAs) that penetrate cell membranes and protect cells from ice crystal formation by reducing intracellular ice formation during freezing^27–29. Another P-CPA, dimethylformamide (DMF), has been successfully applied in sperm cryopreservation across various avian^{17,20,30–34} and mammalian species^35,36. However, to our knowledge, it has not yet been tested for penguin sperm cryopreservation. A second group of cryoprotective compounds, non-permeating cryoprotective agents (N-CPAs), includes sugars that act extracellularly to mitigate osmotic stress during freezing and thawing^{27,29,32,37–39}. Among these, trehalose (α-D-glucopyranosyl-(1→1)-α-D-glucopyranoside) has been combined with different P-CPAs in multiple species, demonstrating its capacity to enhance post-thaw sperm survival^{28,31,32,40,41}. Trehalose occurs naturally in many non-mammalian organisms, where it protects against freezing and desiccation, enabling survival under extreme environmental conditions⁴². Its protective role is explained by the vitrification and water replacement hypotheses: it stabilizes cell membranes, inhibits ice crystal formation, and increases the glass transition temperature of solutions, thereby reducing cryoinjury during frezing⁴². Despite these benefits, trehalose has not been investigated in penguin sperm cryopreservation.

To our knowledge, flow cytometry has previously been applied only to fresh penguin semen, with our earlier study providing the first such analysis in African penguins⁴³. Building on that foundation, we employed flow cytometric assessment in the present study to provide a comprehensive evaluation of functional sperm parameters in frozen-thawed samples of this species.

The objectives of this study were to determine the effectiveness of DMF as a permeating cryoprotectant, assess whether its combination with the non-permeating agent trehalose enhances post-thaw sperm quality, and to establish a foundation for a refined, species-specific sperm cryopreservation protocol for the African penguin.

Results

A total of 27 ejaculates were collected from 10 individual male African penguins. The mean total ejaculate volume was 42 µl, ranging from 10 to 80 µl. The mean sperm concentration was 106.4 × 10⁶ sperm/ml (range: 22.49–664.29 × 10⁶ sperm/ml), and the mean total sperm count per ejaculate was 3.39 × 10⁶ sperm (range: 1.00–10.63 × 10⁶ sperm). Basic semen characteristics of fresh samples are presented in Table 1.

Table 1.

Fresh semen variables for African Penguin Spheniscus demersus. Data presented as mean ± SD and median (range), (n = 27).

Semen characteristic	Descriptive statistics
Volume (µl)	Mean ± SD	42.22 ± 21.89
	Median (range)	40 (10–90)
	Vc (%)	51.85
Sperm concentration (×106/ml)	Mean ± SD	106.40 ± 162.86
	Median (range)	60.99 (22.49–664.29)
	Vc (%)	153.06
Sperm in ejaculate (×106)	Mean ± SD	3.39 ± 2.69
	Median (range)	2.42 (1–10.63)
	Vc (%)	79.35
Total motility (%)	Mean ± SD	42.25 ± 28.92
	Median (range)	28.8 (6.6–87.2)
	Vc (%)	68.45
Progressive motility (%)	Mean ± SD	16.80 ± 14.34
	Median (range)	15.2 (1–44.8)
	Vc (%)	85.36
Rapid (%)	Mean ± SD	42.67 ± 16.90
	Median (range)	41.1 (13.4–77.6)
	Vc (%)	39.61
Slow (%)	Mean ± SD	21.77 ± 14.62
	Median (range)	14.3 (4.7–49.4)
	Vc (%)	67.16
VCL (µm/s)	Mean ± SD	81.53 ± 16.91
	Median (range)	79.64 (50.14–114.05)
	Vc (%)	20.74
VSL (µm/s)	Mean ± SD	41.08 ± 10.03
	Median (range)	39.29 (21.17–65.56)
	Vc (%)	24.42
VAP (µm/s)	Mean ± SD	51.43 ± 9.91
	Median (range)	50.13 (31.14–73.83)
	Vc (%)	19.27
LIN (%)	Mean ± SD	54.42 ± 8.94
	Median (range)	53.36 (34.75–76.64)
	Vc (%)	16.43
STR (%)	Mean ± SD	79.88 ± 6.49
	Median (range)	80.58 (67.36–94.66)
	Vc (%)	8.12
WOB (%)	Mean ± SD	65.53 ± 6.51
	Median (range)	66.59 (48.85–79.91)
	Vc (%)	9.93
ALH (µm)	Mean ± SD	6.31 ± 1.70
	Median (range)	2.89 (2.89–9.61)
	Vc (%)	26.94
BCF (Hz)	Mean ± SD	32.78 ± 13.22
	Median (range)	28.96 (18.53–64.64)
	Vc (%)	40.33

SD = standard deviation, VC = variation coefficient, presented as percentage, VCL = curvilinear velocity, VSL = straight-line velocity, VAP = average path velocity, LIN = linearity, STR = straightness, WOB = wobble, ALH = amplitude of lateral head displacement, BCF = beat cross frequency.

Results of all viability and motility assessments for frozen–thawed samples are summarized in Table 2. Viability assessed by eosin–nigrosin staining did not differ significantly between groups. Total motility (MOT) was significantly higher in the control group compared to the T500 group (p < 0.05), with no significant differences observed in progressive motility (PROG), straightness (STR), or wobble (WOB). The T500 group showed a significantly higher proportion of rapid sperm and increased kinematic parameters (curvilinear velocity – VCL, straight-line velocity – VSL, average path velocity – VAP, amplitude of lateral head displacement – ALH, and beat cross frequency – BCF) compared to the control group (p < 0.05).

Table 2.

Semen variables in frozen – thawed African Penguin samples in two tested groups. Data are presented as mean ± SD and median (range), sample size (n) = 27.

Semen characteristic	Descriptive statistic	Group		p-value	Effect size
		Control	T500
ViabilityA (%)	Mean ± SD	31.07 ± 14.29	32.74 ± 18.14	0.587	0.0764
	Median (range)	28 (13–65)	31 (0–72)
Total motility (%)	Mean ± SD	7.76 ± 7.02	4.21 ± 11.67	0.00186*	0.601
	Median (range)	5.3 (0.7–25.4)	1.9 (0.0–62.2)
Progressive motility (%)	Mean ± SD	0.55 ± 0.45	0.40 ± 0.37	0.265	0.225
	Median (range)	0.4 (0.0–1.9)	0.4 (0.0–1.3)
Rapid (%)	Mean ± SD	13.59 ± 9.83	22.10 ± 15.43	0.0089*	0.502
	Median (range)	9.1 (0.0 33.3)	25.0 (0.0–47.1)
Slow (%)	Mean ± SD	7.37 ± 6.70	3.59 ± 11.67	0.00071*	0.690
	Median (range)	4.5 (1.0–24.8)	1.3 (0.0–61.8)
VCL (µm/s)	Mean ± SD	69.50 ± 28.31	88.00 ± 36.89	0.028*	0.421
	Median (range)	58.19 (35.61–133.60)	90.81 (0.00–160.26)
VSL (µm/s)	Mean ± SD	25.46 ± 8.13	31.95 ± 12.91	0.014*	0.467
	Median (range)	24.02 (15.62–42.69)	34.82 (0.00–49.06)
VAP (µm/s)	Mean ± SD	35.62 ± 12.80	45.11 ± 15.53	0.00645*	0.513
	Median (range)	31.96 (20.79–72.06)	47.15 (0.00–67.26)
LIN (%)	Mean ± SD	44.12 ± 8.19	44.21 ± 15.53	0.413	0.162
	Median (range)	46.26 (23.92–56.94)	49.19 (0.00–66.39)
STR (%)	Mean ± SD	72.96 ± 8.16	68.96 ± 18.89	0.786	0.0555
	Median (range)	75.78 (51.85–83.85)	73.85 (0.00–90.26)
WOB (%)	Mean ± SD	56.45 ± 7.09	56.88 ± 15.27	0.313	0.199
	Median (range)	57.16 (37.64–68.98)	61.06 (0.00–75.72)
ALH (µm)	Mean ± SD	5.95 ± 2.62	8.14 ± 2.47	0.00325*	0.55
	Median (range)	4.64 (3.00–11.23)	8.51 (0.00–11.02)
BCF (Hz)	Mean ± SD	63.67 ± 35.01	80.55 ± 38.30	0.0276*,a	0.370
	Median (range)	54.63 (25.25–165.88)	79.39 (0.00–158.70)

p – value is for the Wilcoxon test for paired samples.

SD = standard deviation, ^A eosin – nigrosine staining assessment, VCL = curvilinear velocity, VSL = straight-line velocity, VAP = average path velocity, LIN = linearity, STR = straightness, WOB = wobble, ALH = amplitude of lateral head displacement, BCF = beat cross frequency, * values statistically significant (p < 0.05), ^a one-sided test.

The features that most clearly differentiated the control group from the T500 group in relation to the first principal component were: slow sperm, total motility (both higher in the control group), VCL, ALH, VAP, BCE, rapid, VSL (all elevated in T500 group). In the context of the second principal component, the most important features were WOB, LIN, and STR, although the visualization of their elevated levels in the control group is not as distinct as the differentiation indicated by the first principal component (Table 3; Fig. 1).

Table 3.

Eigenvalues, percent of cumulative explained variance and loadings of first five principal components of PCA of semen variables in frozen – thawed African Penguin samples.

	PC1	PC2	PC3	PC4	PC5
Eigenvalue	5.28	3.51	1.28	0.82	0.82
Cumulative variance percentage	40.59	67.62	77.51	83.80	90.08
Loadings
Viability (%)	0.23	-0.51	-0.22	0.49	0.55
Total motility (%)	0.60	-0.57	-0.40	-0.34	0.16
Progressive motility (%)	-0.09	-0.53	-0.55	0.42	-0.30
Rapid (%)	-0.81	-0.11	-0.29	0.10	-0.13
Slow (%)	0.63	-0.55	-0.33	-0.38	0.17
VCL (µm/s)	-0.77	0.33	-0.36	-0.25	0.16
VSL (µm/s)	-0.90	-0.29	-0.13	-0.02	-0.13
VAP (µm/s)	-0.93	0.01	-0.25	-0.13	0.00
LIN (%)	-0.31	-0.87	0.34	0.01	-0.08
STR (%)	-0.18	-0.87	0.18	-0.19	-0.14
WOB (%)	-0.41	-0.79	0.36	-0.03	-0.01
ALH (µm)	-0.91	0.12	-0.07	-0.10	0.23
BCF (Hz)	-0.64	-0.04	0.30	0.04	0.48

PC – Principal component, VCL = curvilinear velocity, VSL = straight-line velocity, VAP = average path velocity, LIN = linearity, STR = straightness, WOB = wobble, ALH = amplitude of lateral head displacement, BCF = beat cross frequency.

Fig. 1.

PCA scatterplot of semen variables in frozen – thawed African penguin samples; ellipses correspond to 95% confidence intervals. VCL = curvilinear velocity, VSL = straight-line velocity, VAP = average path velocity, LIN = linearity, STR = straightness, WOB = wobble, ALH = amplitude of lateral head displacement, BCF = beat cross frequency.

Flow cytometry analysis are summarized in Table 4. The percentage of apoptotic spermatozoa was significantly lower in T500 group compared to the control (p < 0.001). Similarly, the percentage of spermatozoa with damaged acrosomes was significantly lower in the T500 group (p < 0.05). In contrast, the control group showed significantly higher proportion of sperm with high mitochondrial membrane potential (HMMP) (p < 0.001) and an elevated DNA fragmentation index (DFI) (p < 0.05). No significant differences were observed between groups in the percentage of live spermatozoa with intact plasma membranes or in the proportion of spermatozoa exhibiting high DNA stainability (HDS).

Table 4.

Sperm characteristics assessed with flow cytometry in frozen – thawed African Penguin semen samples in two tested groups. Data are presented as mean ± SD and median (range); sample size (n) = 27.

Sperm characteristic	Descriptive statistic	Group		p-value	Effect size
		CONTROL	T500
Live (%)	Mean ± sd	32.58 ± 21.56	28.45 ± 19.48	0.225	0.236 (small)
	Median (range)	29.60 (5.73–90.73)	26.00 (3.90–92.84)
Apoptotic (%)	Mean ± sd	15.39 ± 13.17	8.77 ± 11.34	4.92 × 10− 7*	0.832 (large)
	Median (range)	13.48 (0.71–59.02)	4.76 (0.42–53.92)
Damaged acrosome (%)	Mean ± sd	3.99 ± 5.74	2.35 ± 2.36	0.0322*	0.358 (moderate)
	Median (range)	1.57 (0.14–22.10)	1.54 (0.06–11.40)
HMMP (%)	Mean ± sd	6.42 ± 5.39	2.81 ± 3.09	0.00092*	0.610 (large)
	Median (range)	4.91 (0.91–21.31)	1.55 (0.00–12.05)
DFI (%)	Mean ± sd	2.72 ± 2.81	1.27 ± 1.78	0.0013*	0.615 (large)
	Median (range)	1.73 (0.00–11.44)	0.44 (0.00–6.56)
HDS (%)	Mean ± sd	20.93 ± 18.63	17.76 ± 11.67	0.611	0.102 (small)
	Median (range)	14.56 (2.52–76.16)	13.78 (6.32–59.48)

p – value is for the Wilcoxon test for paired samples; SD = standard deviation, HMMP = High Mitochondrial Membrane Potential, DFI = DNA Fragmentation Index, HDS = High DNA Stainability, * values statistically significant (p < 0.05).

Principal Component Analysis (PCA) indicated that the most pronounced differences between the analyzed groups were associated with the DFI and HDS, both of which exhibited higher levels in the control group (Table 5; Fig. 2).

Table 5.

Eigenvalues, percent of cumulative explained variance and loadings of first five principal components of PCA of sperm characteristics assessed with flow cytometry in frozen – thawed African Penguin semen samples.

	PC1	PC2	PC3	PC4	PC5
Eigenvalue	2.35	1.22	0.98	0.63	0.54
Cumulative variance percentage	39.17	59.48	75.89	86.46	95.40
Loadings
Live (%)	-0.6	0.41	-0.10	-0.55	0.33
Apoptotic (%)	-0.53	0.35	0.68	0.26	-0.16
Damaged acrosome (%)	-0.61	-0.53	-0.07	-0.34	0.47
HMMP (%)	-0.47	-0.67	-0.32	-0.30	-0.34
DFI (%)	-0.72	-0.16	0.57	-0.05	-0.16
HDS (%)	-0.73	-0.42	0.29	0.25	-0.20

HMMP = High Mitochondrial Membrane Potential, DFI = DNA Fragmentation Index, HDS = High DNA Stainability.

Fig. 2.

PCA scatterplot of sperm characteristics assessed with flow cytometry in frozen – thawed African penguin semen samples; ellipses correspond to 95% confidence intervals. HMMP = High Mitochondrial Membrane Potential, DFI = DNA Fragmentation Index, HDS = High DNA Stainability.

Correlation analyses showed that, in the control group, HMMP was significantly and positively correlated with MOT and PROG (Table 6). These associations were not observed in the T500 group. In both groups, viability assessed by eosin-nigrosin staining was significantly and positively correlated with the percentage of live spermatozoa determined by flow cytometry using SYBR/PI staining. The strength of these correlations did not differ significantly between groups.

Table 6.

Spearman’s rank correlation analysis of African Penguin chosen spermatozoa parameters.

	Viability1	Total motility	Progressive motility	Live2	HMMP
Viability1	1	0.41	0.26	0.8*	0.09
Total motility	0.56*	1	0.7*	0.33	0.69*
Progressive motility	0.33	0.8*	1	0.21	0.61*
Live2	0.6*	0.37	0.13	1	0.11
HMMP	0.23	0.44	0.43	0.23	1

Correlation coefficients calculated within the control group are presented in the upper triangle of the matrix, while those calculated within the T500 group are shown in the lower triangle. * correlation coefficients that were statistically significantly different from zero. The corresponding correlation coefficients were not statistically significantly different.

¹ light microscopy assessment with eosin – nigrosine staining; ² flow cytometry assessment with SYBR/PI staining; HMMP = High Mitochondrial Membrane Potential.

Discussion

To the best of our knowledge, this is the first study to investigate the use of DMF as a permeating cryoprotectant, and its combination with trehalose, a non – permeating cryoprotectant, for semen cryopreservation in any penguin species. Previous studies on penguin semen cryopreservation have relied exclusively on permeating cryoprotectants^21,22,24,25, which, when not properly dosed, may exert toxic effects on sperm cells^27,38. The consistently modest outcomes observed in frozen-thawed semen across penguin species highlight the limitations of using these agents alone^21,22,24,25. Given that cryopreservation inevitably reduces semen quality and that post-thaw values in penguins remain lower than in most other avian species^27,31,32, we aimed to assess whether combining permeating and non-permeating cryoprotectants could provide synergistic benefits and improve post-thaw sperm functionality.

In our study on African penguins, the use of DMF alone yielded post-thaw viability and motility values within expected ranges for the species^22,24 but lower than values typically reported for other birds^27,31,32. For instance, in Thai red junglefowl (Gallus gallus domesticus), 6% DMF improved post-thaw viability and motility compared to 6% or 9% DMA or MA (N-methylacetamide)³¹. Similarly, 6% DMF has been used effectively in chickens and guinea fowl (Numida meleagris), producing fertility rates between 79% and 90% in some chicken breeds. In ducks, 8% DMF was superior to 6%²⁷, while in ganders, 6% DMF also improved cryopreservation outcomes³⁴. In contrast, the values obtained in our study closely matched those previously reported for other penguin species^21,22,24,25, suggesting a species-specific response pattern likely reflecting intrinsic cryobiological characteristics.

Both treatment groups in our study showed no significant differences in viability, whether assessed via eosin–nigrosin staining or SYBR/PI fluorescent staining (evaluated as the percentage of live spermatozoa with intact plasma membranes). Viability rates were comparable to those reported by Marti-Colombas et al.²⁴, who found post-thaw viability of 32% with DMSO and 30.5% with DMA in African penguins. This consistency suggests that African penguin sperm maintains a relatively stable viability profile in frozen-thawed samples, regardless of the cryoprotectant or assessment method. Correlation analyses confirmed a strong association between the two viability assessment techniques, supporting the reliability of simpler morphological staining methods such as eosin–nigrosin. This is particularly relevant for wildlife reproductive research, where rapid, cost-effective assessments are often required under field conditions without advanced laboratory facilities.

Regarding total motility (MOT), the control group treated with 6% DMF alone performed significantly better than the T500 group, which received a lower DMF concentration (3%) combined with trehalose. A similar trend was observed for high mitochondrial membrane potential (HMMP), where the control group again showed significantly better outcomes. Correlation analysis revealed a strong positive correlation between HMMP and MOT (r = 0.69) and between HMMP and progressive motility (PROG) (r = 0.61) within the control group only. Although mitochondrial potential is known to be associated with sperm motility, the strength and nature of this relationship depend on multiple factors, and relevant avian data remain limited, highlighting the need for further study²⁰. Overall, our findings indicate that 6% DMF may be more effective as a permeating cryoprotectant for preserving motility than a reduced concentration combined with trehalose.

These results contrast with those reported in chickens, where the combined use of DMF and trehalose improved not only MOT and PROG but mitochondrial activity as well³². Nonetheless, when compared with other permeating cryoprotectants tested in African penguins, our MOT values fell within a similar range²⁴. They exceeded the 3.1% MOT observed with DMA but remained below the 13.3% MOT reported for DMSO by Marti-Colombas et al.²⁴. Similarly, although PROG did not differ significantly between our treatment groups, both values were higher than the 0% recorded for DMA, yet lower than the 1.3% achieved with DMSO in the same study²⁴. Interestingly, despite the lower MOT in the T500 group, several kinematic parameters, specifically curvilinear velocity (VCL), straight-line velocity (VSL), and average path velocity (VAP), were significantly higher in this group than in the control. These findings parallel those of Mosca et al.⁴⁴, who reported that trehalose selectively improved certain motion characteristics of cryopreserved chicken sperm. The enhancement of VSL may hold particular biological relevance, as this parameter reliably reflects sperm velocity and correlates with fertility outcomes in birds⁴⁴. The elevated VSL in T500 samples therefore suggests that, despite lower MOT, sperm in this group may possess enhanced fertilizing potential. Moreover, the addition of trehalose increased VCL and beat cross frequency (BCF), both indicators of the sperm’s capacity to traverse the female reproductive tract and interact with the oocyte^45,46. According to Mosca et al.⁴⁴, these parameters are linked to zona pellucida penetration ability in mammalian sperm and may similarly reflect post-thaw fertilization competence in avian species. Conversely, linearity (LIN) and wobble (WOB), parameters associated with directional motility, did not improve following trehalose addition. This agrees with observations in chicken sperm⁴⁴ and may explain why PROG remained unaffected in our study despite observed enhancements in other kinematic traits.

PCA further supported these findings by identifying motility-related variables as key contributors to the differences between treatment groups.

These results provide a basis for refining cryopreservation protocols to better preserve sperm motility, which is essential for maintaining viability after deposition in the cloaca and for subsequent storage within the female reproductive tract¹⁶. As cryopreserved semen is intended for artificial insemination to support the restoration of genetic diversity in declining populations^16,26, improving post-thaw motility remains a key priority.

During cryopreservation, the acrosome of avian spermatozoa is highly vulnerable to osmotic stress, ice crystal formation, and oxidative stress, which can compromise its integrity, induce premature acrosome reactions or structural disruption, and ultimately impair fertilization ability^27,47. Although our flow cytometry results showed that the T500 group had a significantly lower proportion of acrosomal damage than the control, both groups exhibited relatively low levels of live sperm with damaged acrosomes – 3.99% in the control and 2.35% in the T500 group – which represents a favorable outcome for frozen-thawed semen in this species⁴³. Comparable values were reported by Partyka et al.³⁴ in ganders, where only 3.8% of live sperm displayed ruptured acrosomes with DMF use. Moreover, studies in chickens⁴⁸, geese⁴⁹, and other species such as buffalo, boars, and dogs³⁸ have demonstrated that trehalose effectively protects the acrosome from cryo-induced damage when combined with various P-CPAs. Similarly, combining DMF and trehalose improved acrosomal integrity in chicken spermatozoa³². Together, these findings indicate that while DMF alone ensures satisfactory preservation of acrosomal structure, the addition of trehalose offers a clear advantage in maintaining acrosomal integrity during cryopreservation.

Apoptotic-like changes are frequently observed in post-thaw sperm due sub-lethal cryodamage^50,51, making their assessment via flow cytometry particularly relevant. In our study, the T500 group exhibited a significantly lower percentage of apoptotic sperm, indicating that trehalose supplementation may enhance post-thaw viability. A reduced proportion of apoptotic cells suggests limited activation of programmed cell death pathways during the freezing and thawing. To our knowledge, no published data exists on apoptotic-like changes in either fresh or cryopreserved semen of wild avian species, complicating direct comparisons. However, in our previous study on fresh African penguin semen, we reported 7.76% apoptotic sperm⁴³, comparable to values in chickens (8.6%)⁴⁸, and similar to those observed in the T500 group. These findings align with studies showing the anti-apoptotic properties of trehalose in spermatogonial stem cells cryopreervation⁵². Given that apoptosis indicates sub-lethal cryodamage, the reduced apoptotic rate in the T500 group suggests a more favorable cellular response to the cryopreservation protocol, potentially associated with improved post-thaw sperm function and fertilizing ability.

Considering the central role of DNA in reproduction and the multiple cryo-induced stressors that may compromise its integrity, including impaired antioxidant systems, ice crystal formation, osmotic imbalance, and activation of apoptotic pathways^53,54, we also analyzed DNA fragmentation. In ganders, 6% DMF preserved DNA integrity post-thaw without significantly affecting the DNA fragmentation index (DFI)³⁴. Similar observations were made in chickens and Barbary partridges (Alectoris barbara), where cryopreservation did not significantly alter DFI⁵⁵. Conversely, other studies in chickens showed substantial increase in DNA fragmentation after cryopreservation, with DFI rising from 6.2% in fresh semen to 19.8% when 6% DMA was used⁵⁶. In our study, the T500 group showed significantly lower DNA fragmentation (DFI = 1.27%) compared with the control group, which also maintained a low level of DNA damage (DFI = 2.72%). Compared with our previously published data on fresh African penguin semen (DFI: 1.83%)⁴³, both cryopreserved groups demonstrated satisfactory DNA preservation. This contrasts sharply with the higher DNA fragmentation reported in African penguin sperm cryopreserved with DMSO (DFI: 14.8%) or DMA (DFI: 12.4%) and in Gentoo penguins (16.2% and 11.3%, respectively)²⁴. These results suggest that DMF may serve as a more effective permeating cryoprotectant for preserving DNA integrity in penguin sperm compared to DMSO or DMA. Moreover, trehalose appears to enhance this protective effect, likely through extracellular osmotic stabilization.

Although intra-species variability was not directly examined, differences among treatment groups indicate that future studies could benefit from integrating individual semen characteristics to optimize cryopreservation outcomes. For example, males with higher baseline acrosomal damage or DNA fragmentation might benefit from trehalose-supplemented protocols, whereas samples with reduced mitochondrial potential could respond better to higher DMF concentrations. Such personalized approaches, however, require further investigation and validation.

Our findings provide valuable insight into how different cryoprotectant combinations, particularly DMF and trehalose, affect post-thaw sperm quality parameters in African penguin, a species of critical conservation concern. These results reinforce that cryopreservation outcomes in birds vary not only across species but also with the type and concentration of cryoprotectants used.

Conclusions

African penguin sperm maintains a relatively stable post-thaw viability regardless of the cryoprotectant or assessment method, supporting the use of simple staining techniques like eosin–nigrosin for reliable and practical evaluation in wildlife research.

In our study, 6% DMF not only better preserved motility and mitochondrial function but also contributed to maintaining DNA integrity, while the addition of trehalose further enhanced this effect and improved acrosomal integrity and apoptosis resistance. These findings emphasize that cryoprotectant selection should be tailored to the specific cellular responses of the target species.

Overall, our results provide a strong foundation for developing more effective, species-appropriate cryopreservation protocols for African penguins, thereby supporting the establishment of avian genetic resource banks and enhancing conservation strategies for critically endangered bird species.

Methods

Animals

The African penguins involved in this research were kept in an outdoor habitat at Zoo Wrocław (Wrocław, Poland), which includes a 900 m² beach area and a pool holding 2,460 m³ of water. Their daily diet consisted of frozen-thawed capelin, herring, and sprat, with an added nutritional supplement – Small Bird Supplement (Mazuri^® Exotic Nutrition, PMI Nutrition International LLC, USA). The study was reviewed and approved by the Animal Welfare Committee of the Wrocław University of Environmental and Life Sciences (Approval No: 3 K.2022) and was carried out in compliance with the ethical standards set by the European Association of Zoos and Aquariums (EAZA) Code of Ethics.

Semen collection

Semen collection was conducted during the breeding season, from July to December 2024, involving 10 sexually mature males that exhibited typical reproductive behaviors such as courtship displays and nest building. Prior to each semen sampling, individual males were temporarily separated from the main colony. The procedure was carried out by two trained researchers using the dorso-abdominal massage technique described by Burrows and Quinn⁵⁷, with adaptations made for this species⁴³. One researcher, skilled in penguin handling, restrained the bird by placing it on a flat, towel-covered surface with its legs extended. One hand was used to immobilize the head, covering both eyes and beak to prevent biting, while the other hand secured one wing against the bird’s body and stabilized the opposite wing against the handler’s torso, ensuring complete immobilization. The second researcher, experienced in semen collection, performed gentle, alternating strokes along the bird’s back and abdomen to stimulate ejaculation. This was followed by light pressure applied to the cloacal region to facilitate semen release. The ejaculates were collected using a capillary tube connected to a suction device and subsequently transferred into 1.5 ml Eppendorf tubes. Semen volume was measured via micropipette. In cases of low-volume samples, Dulbecco’s Modified Eagle Medium (DMEM, cat. no. D5921, low glucose, without phenol red)⁵⁸ was used for dilution to prevent desiccation. All samples were analyzed within one hour of collection. After each procedure, the males were returned to their main enclosure.

Semen processing for cryopreservation

All collected semen samples were processed individually. Sperm motility parameters, including total motility (MOT, %), progressive motility (PROG, %), and detailed kinematic variables – curvilinear velocity (VCL, µm/s), straight-line velocity (VSL, µm/s), average path velocity (VAP, µm/s), linearity (LIN, %), straightness (STR, %), wobble (WOB, %), amplitude of lateral head displacement (ALH, µm), and beat cross frequency (BCF, Hz) – were evaluated using a computer-assisted sperm analysis (CASA) system, CEROS II (Hamilton Thorne Biosciences, MA, USA). Progressive motility was defined as STR > 75% and VAP > 50 μm/s. Head detection was limited to 2–120 μm². A volume of 4 µl from each semen sample was loaded into pre-warmed Leja chamber slides (Leja Products B.V., Nieuw Vennep, The Netherlands) and analyzed on a stage maintained at 39 °C. For each sample, five randomly chosen microscopic fields were examined at a rate of 60 frames/s to derive the motility and kinematic data.

Because of varying sperm concentrations and the lack of information available on insemination dose requirements in African penguins, only samples containing at least 1 million spermatozoa per ejaculate were selected for cryopreservation. Each qualifying sample was divided into two equal aliquots, each containing approximately 0.5 to 1 million spermatozoa. One aliquot was diluted 1:1 with Lake and Ravie diluent⁵⁹ (hereinafter referred to as control), while the other was diluted 1:1 with the same medium supplemented with trehalose at a concentration of 500 mM (hereinafter referred to as T500). Both were chilled for 1 h at 5 °C. Following this, 6% DMF was added to the control, and 3% DMF to the T500 test group. The samples were then equilibrated for 10 min at 5 °C. Each preparation (~ 0.7 × 10⁶ sperm/straw) was loaded into 0.25 ml plastic French straws (IMV Technologies, France) and sealed using a Biotherm™ HSA-1 straw 2 sealer set to 13 °C, as per manufacturer’s instructions. The straws were then placed horizontally on a rack suspended 5 cm above the surface of liquid nitrogen for 15 min to allow controlled freezing to approximately − 140 °C. Subsequently, the straws were plunged into liquid nitrogen (–196 °C), stored in plastic goblets, and maintained in a cryotank filled with liquid nitrogen until thawing. A total of 54 (27 per each group) straws were frozen. The straws were thawed in a water bath at 37 °C for 30 seconds¹⁷, after which their contents were transferred into Eppendorf tubes for post-thaw sperm quality assessment.

Thawed samples assessment

Sperm viability was assessed using the eosin-nigrosin staining method, based on plasma membrane integrity, where spermatozoa staining pink were classified as nonviable, and unstained (white) cells were considered viable. A small volume of semen (2–3 µl) was mixed with 5 µl of eosin-nigrosin stain on a microscope slide, air-dried, and subsequently examined under a light microscope using an oil immersion objective (1250× magnification). A total of 200 spermatozoa per slide were evaluated to calculate the percentage of live and dead cells.

Sperm motility parameters were assessed using the same methodology as applied to the samples prior to cryopreservation.

Flow cytometry analysis was conducted as previously described in Borecki et al.⁴³, using a Guava EasyCyte 5 cytometer (Merck KGaA, Darmstadt, Germany), equipped with a 488 nm Argon ion laser for excitation. Data acquisition was performed with GuavaSoft™ 3.1.1 software. Samples were diluted to a final sperm concentration of 5 × 10⁶ sperm/ml, and 10,000 events were recorded per sample. Non-sperm events were gated out based on forward and side scatter characteristics. The sheath flow rate was maintained at 0.59 µl/min. Fluorescent signal detection was performed using the following filters: FL1 (525/30 nm) for green fluorescence (SYBR-14, PNA-Alexa Fluor^® 488, monomeric JC-1, and YO-PRO-1), FL2 (583/26 nm) for orange fluorescence (JC-1 aggregates), and FL3 (695/50 nm) for red fluorescence (propidium iodide – PI). A compensation setting of 7.9% was applied to correct SYBR-14 spectral overlap into the PI channel. All data, collected in logarithmic mode for FSC, SSC, and fluorescence, were stored in FCS files and exported as Excel spreadsheets (.xlsx) for further analysis.

Percentage of live spermatozoa was estimated from plasma membrane integrity assessment using SYBR-14 and PI dual staining (Live/Dead Sperm Viability Kit, cat. no. L7011, Invitrogen, USA). A 300 µl aliquot of diluted semen was incubated with 5 µl SYBR-14 (0.33µM) for 10 min at room temperature in the dark. Then, 3 µl PI (23µM) was added 5 min before cytometric analysis. Cells emitting green fluorescence (SYBR-14⁺/PI⁻) were considered viable with intact membranes, red fluorescent cells (SYBR-14⁻/PI⁺) as nonviable, and double-stained cells (SYBR-14⁺/PI⁺) were interpreted as dying or membrane-compromised. The population of viable cells with intact membranes (hereinafter referred to as Live) was used for comparisons between the tested groups.

Acrosome integrity was determined using PNA-Alexa Fluor^® 488 conjugate (cat. no. L21409, Life Technologies, USA). Sperm samples were diluted (300 µl), stained with 10 µl of PNA (1 µg/ml), and incubated for 5 min at room temperature in the dark. After staining, samples were centrifuged (600 g, 3 min), the supernatant was discarded, and the pellet was resuspended in 300 µl of DMEM. PI (3 µl) was added immediately prior to cytometric analysis. Based on PI and PNA signals, four populations were identified: live sperm with intact acrosome (PI⁻/PNA⁻), live with damaged acrosome (PI⁻/PNA⁺), dead with intact acrosome (PI⁺/PNA⁻), and dead with damaged acrosome (PI⁺/PNA⁺). The population of live sperm with damaged acrosome was used for comparisons between the tested groups.

Mitochondrial membrane potential (MMP) was evaluated using JC-1 and PI (cat. no. T3168, Life Technologies, USA). Diluted samples (300 µl) were incubated with 0.67 µl JC-1 (6.7µM) for 20 min at 37 °C in the dark. Prior to analysis, 3 µl PI was added. Dead cells (PI⁺) were excluded via FL3/FL2 gating, and live cells were analyzed based on FL2/FL1 signals. Orange fluorescence indicated high mitochondrial membrane potential (HMMP) (JC-1 aggregates), while green fluorescence corresponded to low mitochondrial membrane potential (LMMP) (JC-1 monomers).

Apoptotic status was assessed using the YO-PRO-1/PI Apoptosis Kit (cat. no. Y3603, Life Technologies, USA). One µl of YO-PRO-1 was added to 300 µl of diluted semen (25nM) and incubated at room temperature for 10 min in the dark. Subsequently, 3 µl PI was added 5 min before analysis. YO-PRO-1⁺/PI⁻ cells were considered early apoptotic, PI⁺/YO-PRO-1⁻ as dead, dual-positive cells as apoptotic and dead, and dual-negative cells as viable and non-apoptotic. Populations of YO-PRO-1⁺/PI⁻ apoptotic spermatozoa were used for comparisons between the tested groups.

Chromatin integrity was evaluated using acridine orange (AO; cat. no. A1301, Life Technologies, USA). Fifty µl of diluted semen was mixed with 200 µl of acid lysis solution (0.1% Triton X-100, 0.15 M NaCl, 0.08 M HCl, pH 1.4), followed by 600 µl of AO solution (6 µg/ml in buffer: 0.1 M citric acid, 0.2 M Na₂HPO₄, 1 mM EDTA, 0.15 M NaCl, pH 6). Spermatozoa emitting predominantly green fluorescence represented intact, double-stranded DNA, while cells exhibiting red or decreased green fluorescence indicated fragmented or denatured DNA. The percentage of sperm with DNA fragmentation index (%DFI) and high DNA stainability (%HDS) was calculated. %HDS was defined by gating above the main intact DNA population.

Statistical analyses

Statistical analysis was performed using R 4.4.1⁶⁰. Statistical inference was conducted at a significance level of 0.05.

To assess the statistical significance of differences in semen parameters between the examined groups (control and T500), the Wilcoxon signed-rank test for paired observations was applied. A non-parametric method was chosen due to the lack of normal distribution in the analyzed variables. Effect sizes were calculated for each of the statistical tests performed⁶¹.

Spearman’s rank correlation analyses with FDR correction were performed to determine whether mitochondrial activity is associated with motility parameters in the tested groups. The same statistical approach was applied to assess the comparability between viability results obtained using eosin – nigrosine staining under light microscopy and those obtained via SYBR/PI staining in flow cytometric analyses, expressed as the percentage of live spermatozoa with intact plasma membranes. The significance of each correlation coefficient was assessed, as well as the statistical significance of differences between corresponding correlation coefficients calculated within the analyzed groups. Correlation coefficients were considered statistically significantly different if their 95% confidence intervals did not overlap. This part of the analysis was performed using the psych library⁶².

To visualize and facilitate the identification of the features that most distinctly differentiate the control group from the T500 group in terms of semen parameters, a Principal Component Analysis (PCA) was performed. The analysis considered principal components with eigenvalues greater than 1 and a cumulative explained variance exceeding 75%. PCA was conducted separately for all sperm kinetic parameters and viability and for semen characteristics assessed using flow cytometry. This part of analysis was performed using the ade4⁶³ and factoextra⁶⁴ libraries.

Author contributions

P.B. conceptualization, data collection and analysis, methodology, writing; A.P. supervision, funding acquisition, methodology, formal analysis, writing–review and editing; A.M. statistical analysis; W.N. supervision, funding acquisition. All authors have read and agreed to the published version of the manuscript.

Funding

PhD student in the 6th edition of the Implementation Doctorate Programme - Ministry of Science and Higher Education. The article is part of a PhD dissertation titled “Development of cryopreservation method for African penguin (Spheniscus demersus) semen to establish a biobank for genomic resources of critically endangered species”, prepared during Doctoral School at the Wrocław University of Environmental and Life Sciences. The APC/BPC is financed by Wrocław University of Environmental and Life Sciences.

Data availability

The datasets used and analyzed during the current study are available from the corresponding author on reasonable request.

Declarations

Competing interests

The authors declare no competing interests.

Ethics

Our study is reported in accordance with ARRIVE guidelines (PLoS Bio 8(6), e1000412,2010​).