Hybridization in Vipers—A Case Study on Mating Between Vipera ammodytes transcaucasiana and V. a. ammodytes in Captivity

Abstract

In the present study, we examine the possibilities of planned generation in snakes by controlling the two most important factors for their estrus—ambient temperature and daylight hours. As a result of controlling these environmental parameters in an increasing gradient until reaching optimal values for the species, we observed copulations in late March and early April between a female Vipera ammodytes transcaucasiana and a male V. a. ammodytes. After three months of “pregnancy”, we obtained viable offspring in early July, which is about two months earlier compared to wild populations. The species used in the experimental setup, in natural conditions, usually produce offspring in late August to early October. Another aspect considered in the publication and followed in the experimental setup was to possibly test if in evolutionary and developmental aspect, both subspecies are closely related and interspecies breeding is possible (which might indicate mutual ancestry). The hybrid individuals were monitored during their entire development from newborns to subadults for pathological traits during development to roll-out crossbreeding incompatibility. In our pilot investigation, no acquired or inherited pathological traits have been observed. The individuals were consistent with feeding and exhibited excellent individual development. Future research coupled with genetic investigation can give valuable insight in the field, whether it is valid to regard the genera as a Vipera ammodytes complex or as different subspecies groups.

1. Introduction

European vipers /genus Vipera/ are distributed almost everywhere in Europe (including a large part of the Mediterranean islands) and in parts of Western and Central Asia, as well as an isolated area in Northwest Africa. The entire genus includes about 33 species and subspecies [1]. The general distribution covers the vertical range from 0 m above sea level and rises to over 3000 m above sea level, and in the horizontal range from the Mediterranean Sea (including the islands) to the Arctic Circle. The ranges of the different species and subspecies overlap in certain areas, creating contact zones and providing conditions for the formation of the so-called lines/zones of hybridization [2,3,4].

According to some authors [3,4], hybridization occurs in areas strongly dominated by natural and/or anthropogenic segregation of the landscape. Of course, there is always the option of non-fertile generations, and in this case the hybrid traits die out with the ontogenetic development of the individual. Guiller et al. [2] argued that different habitats and thermal specializations of vipers can provide a parsimonious explanation for understanding large-scale parapatric distributions, climatic niches and the partitioning of microhabitats into contact zones, leading to limited gene flow. With regard to vipers, the situation appears somewhat different in easternmost Europe (i.e., Ukraine and the Caucasus region), where species differentiation is much more recent (about 4 million years) and complex—especially within the subgenus Pelias [5,6]. Multiple contact zones or overlapping distributions between related species have been described, as introgression has been found in Caucasian vipers [5,6]. A recent study showed that the rare Vipera magnifica and Vipera orlovi were generated by contact and hybridization between two species complexes—Vipera kaznakovi and Vipera renardi [6]. These variable levels of hybridization among Eurasian vipers may reflect different degrees of genetic differentiation between related taxa, caused by recent glacial fluctuations or habitat fluctuations [6]. There are currently two officially recognized hybrid species of Palaearctic vipers—Vipera magnifica [7] and Vipera pontica [1].

In general, the Horned-nosed viper (Vipera ammodytes) can be subdivided into four subspecies with the following distribution—V. a. ammodytes (from Italy and Austria to the western and central parts of the Balkans); V. a. montandoni (eastern and southern Balkans to western Turkey); V. a. meridionalis (southernmost parts of the Balkans); V. a. transcaucasiana (eastern Turkey, Georgia and Armenia) [8]. NW—Italy, Austria, Slovenia, Croatia, Bosnia and Herzegovina; NE—western, central and eastern Serbia, western Bulgaria; MO—Montenegro; SW—Albania and northwestern Greece; SE—southern Serbia, North Macedonia, central and eastern Bulgaria, northern and central Greece, Turkey, Armenia and Georgia; PE—Peloponnese peninsula; CY—Cyclades Islands.

Such a large genetic diversity could be explained in the following way: the intraspecific splits of the species V. ammodytes began in the early Pliocene (about 3.4–4.9 million years ago) when the seas entered the central parts of the Balkans and the formation of refugia (the Rhodope, Pindus, Dinaric massifs) and subsequent settlement of the territories freed from seawater [9]. From what has been said so far, it can be concluded that the CY, PE and MO clades have separated and differentiated in certain areas and as a whole do not affect genetic transfer. The two northern clades extend almost only within the natural boundaries of the nominate subspecies V. a. ammodytes, where areas of overlap are observed in Western and Central Serbia [9]. Of the two southern clades, the southeastern clade is of greater interest, which can be traced from Armenia and Georgia, through Turkey and Greece to Western Bulgaria, and is found in three of the viper subspecies (V. a. trancaucasiana, V. a. meridionalis, V. a. montandoni). In our opinion, as Dyugmedzhiev et al. [10] point out, the territory of Bulgaria is one of the most interesting areas for studying the phylogeny of the species Vipera ammodytes. Here, the boundaries of the ranges of two of the subspecies (V. a. ammodytes and V. a. montandoni) overlap very closely, creating potential areas of hybridization, which creates opportunities for genetic exchange between the NE and SE clades and the appearance of hybrid specimens in the contact zones. Through the subspecies V. a. ammodytes NE clade, present in Bulgaria in the northern and central parts of Western Bulgaria and in the triangle between the Iskar and Malki Iskar rivers, overlaps with the SE/E subclade of the SE clade, creating a polygon for potential interaction. Another such polygon is concluded between the southern slopes of the Stara Planina, the eastern slopes of Ihtimanska Sredna Gora and the western slopes of Saštinska Sredna Gora. In the Kraishte region, one could also look for a contact zone between the NE and SE clades, but in this case the SE would be represented by the SE/S subclade. At the moment, these remain probable zones, as the studies are at their beginning and more samples and information need to be accumulated. The same applies to the Bulgarian territory of the Mesta river valley, but for contact between the SE/E and SE/S subclades (genetically mixed specimens have been found to the south in Northern Greece). Our experiment corresponds with the study of Freitas et al. [7], where the four subspecies of Vipera ammodytes are placed in a single subclade known as Vipera 2 with a single genetic lineage comprising the Vipera ammodytes-transcaucasiana complex.

The goal of the present study was to test some aspects in the mating behavior in three subspecies of Vipera ammodytes: V. a. ammodyres, V a. meridionalis, and V. a. transcaucasiana, when housed together in captivity under modified conditions. We propose that all specimens will be involved in the acts of concurrence and/or courtship, despite representing different subspecies.

2. Material and Methods

2.1. General Information for the Experimental Specimen

In the winter of 2021, at the zoological collection of the “Research and breeding base of Herpetologica Ltd. (Burgas postcode 8000, Bulgaria)”, we formed a combined experimental group of animals including 4 sexually mature vipers—Vipera ammodytes ammodytes (1 male and 1 female), V. a. montandonii (1 male), and V. a. transcaucasiana (1 female). All animals were part of the private collection of Herpetologica Ltd. and were rescued animals with legal documents issued from the Ministry of Environment and Waters of Bulgaria. For the purposes of this experiment, we used a modified animal preparation methodology according to Hristov et al. [11].

2.2. Hibernation

In December 2021 (from 15. 12. 21), the animals were put into a state of hibernation for about 2.5 months (until 01. 03. 22)—the approximate period for wintering of vipers in natural conditions (this is important for these species, because they have post-hibernation gametogenesis [12]). Hibernation was carried out in the dark (for this purpose, lighting in the terrariums was eliminated) and cool conditions (the temperature was gradually lowered to 12–15 °C). The animals were provided with drinking water and a moist substrate (to avoid dehydration).

2.3. Post Hibernation Preparation

On 1 March 2022, the animals began to emerge from hibernation—they were moved to a terrarium (100 × 50 × 50 sm) with a dry substrate and were provided with water. The environment was gradually warmed up to the optimal 25–28 °C and the daylight hours were increased to 8–10 h. Increasing heat and light should be done gradually within 20–30 days. After this adaptive period, a plateau of the two environmental factors (heat and light) was reached. It was up to 12–14 h of daylight and 28–30 °C temperature (which is important for the wear of the fetus by the females).

2.4. Start of the Feeding

Five days (05. 03. 22), after the end of hibernation, the animals were offered food–mice weighing 25–30 g. The animals attacked the prey and fed willingly, at an average temperature of about 20–22 °C, and a light day of about 6 h, and had no problems with swallowing and digesting. After another ten days (15. 03. 22), similar prey was offered again and the same behavior was observed—willingly attacking and swallowing the prey. This was the preparatory phase before the start of the mating ethological complex. Since the establishment of the experimental group, the four animals have not been separated.

2.5. Documentation

Weight measuring was performed using Pocket scale MH—Series *Auto Calibration scale EK 6002; all photographs were taken using a Nikon D3100, lens Nikon DX AF—5 Nikkor 18–55 mm 1:3.5–5.6 G. All data generated were exported and analyzed in Microsoft Excel Version 2019 (Microsoft Corporation)

3. Results

3.1. Mating Behaviour

The male vipers began to show the first signs of heat (at 17. 03. 22.)—they become “nervous” in case their bodies touched any of the other animals, the walls of the terrarium, or objects from the interior. They tried to push subjects away, while simultaneously raising the front part of their body together even during feeding. The controlled and significant environmental factors—light and heat—were already within the limits of 25 °C and a light day of approximately 8 h.

On 22. 03. 22, the male vipers began tournament fights, which lasted until 24. 03. 22 with varying durations and intensities. In between, each of the males tried to copulate with the nominated female. On 24. 03. 22, the winner of the tournament, the male of the nominated subspecies, began to court the female of the same species. Copulation took place on 26. 03. 22 and lasted about 4 h. On 31. 03. 22, during the feeding the animals, surprisingly the tournament fights between the males started again and lasted until 02. 04. 22. The winner was again the male of the nominated subspecies. In a very short period (literally a few hours) it began to court the female Armenian nose-horned viper (Supplement Video S1). On 04. 04. 22, the male copulated with it for about 5 h (Figure 1).

3.2. Post-Mating Period in the Adults

In order to ensure secure place for the females, they were separated into new cages. The males continued to live in the same terrarium and for the same year they did not show sexual activity any more.

3.3. Appearing and Development of the Hybrid Generation

The two separated females ate their usual prey (mice 25–30 g) rhythmically every 10–15 days, gradually reducing the meals and preferring smaller ones (mice 15–20 g). In early June 22, they finally refused food until after the “birth”. The female V. a. ammodytes has never produced offspring. On 4 July 2022, the female V. a. transcaucasiana gave birth to 6 live juveniles—3 males and 3 females (Figure 2)—without releasing any yolk sacs or dead specimens. Within a few hours, all offspring shed their skin. Each of the juvenile specimens was placed in an individual plastic transport box, equipped with holes for passive ventilation and elimination of condensation, lined with kitchen paper. Fresh water and food were provided regularly (

Table 1. Data on the feeding events and feeding ethology in the juvenile V. a. ammodytes × V. a. transcaucasiana hybrids: the numbers represent the single hybrid specimen; the green lines indicate the year; the checked white fields represent the successful feeding processes; the red unchecked fields represent unsuccessful feeding events; checked red field indicate on partly successful feeding process (food was ingested but regurgitated in 2–5 days).

Date		Feeding
2022	1	2	3	4	5	6
14.07	√
21.07	√	√	√
29.07	√	√	√	√	√
06.08	√	√	√	√	√
19.08	√	√	√	√	√	√
26.08	√	√	√	√	√	√
11.09	√	√	√			√
28.09	√		√	√	√	√
08.11	√	√	√			√
21.11	√	√	√		√	√
30.11	√	√	√		√	√
08.12	√	√	√		√	√
20.12	√		√		√	√
31.12	√	√	√		√	√
2023
15.01	√	√	√	√	√	√
29.01	√	√	√	√	√	√
11.02	√	√	√	√	√	√
21.02	√	√	√	√	√	√
03.03	√	√	√	√	√	√
13.03	√	√	√	√	√	√
26.03	√	√	√	√	√	√
06.04	√	√	√	√	√	√
19.04	√	√	√	√	√	√
04.05	√	√	√	√	√	√
26.05	√	√	√	√	√	√
03.06	√	√	√	√	√	√
03.07	√	√	√	√	√	√
16.07	√	√	√	√	√	√
30.08	√	√	√	√	√	√
22.10	√	√	√	√	√	√
04.12	√	√	√	√	√	√
18.12	√	√	√	√	√	√
Date		Feeding
2024	1	2	3	4	5	6
16.01	√	√	√	√	√	√
23.02	√	√	√	√	√	√

).

Initially we waited 10 days before offering food, so that the juveniles could get used to their new habitats (the plastic boxes) and absorb the residual reserves from their intrauterine development. On 14. 07. 22, we offered newborn mice (<3 g) to all juvenile vipers, but only one attacked and swallowed the prey. Until all were fed, food was offered every 7 days. On 06. 08. 22, five out of all six juveniles attacked the prey. From 19. 08. 22, we started offering them food at intervals of 10–17 days. In general, all of the juveniles eat rhythmically and gain weight proportionally (Table 1 and

Table 2. Data on the growth rate in the juvenile V. a. ammodytes × V. a. transcaucasiana hybrids: numbers represent every single specimen; green lines represent the years; light violet lines represent the change in the size of prey (in g); blue represent a growth spurt in summer of 2023.

Date			Weight in g
2022	1	2	3	4	5	6
Born	04.07	04.07	04.07	04.07	04.07	04.07
Sex	♂	♀	♀	♂	♀	♂
Shedding	02.12	24.12	17.12	15.01 23	14.12	03.12
2023	13.02	16.02	14.02	07.03	06.03	13.02
	08.04	28.03	06.04	20.04	18.04	04.04
	17.06	13.06	10.07	17.06	19.06	15.06
08.09	10.68	9.11	10.97	9.46	8.12	5.66
20.09	10.94	9.00	11.40	9.34	7.46	6.02
03.11	11.25	8.35	12.14	8.98	7.31	6.90
21.11	11.19	8.23	12.24	7.50	6.04	6.82
30.11	12.03	8.76	12.74	7.11	6.62	7.74
07.12	12.00	10.48	14.42	6.81	7.07	7.37
20.12	12.48	10.27	14.02	6.23	7.29	7.72
31.12	13.70	9.08	14.73	5.90	7.91	8.74
2023
15.01	16.70	12.23	18.64	10.29	10.83	10.55
29.01	16.62	11.16	18.42	8.48	10.62	11.00
11.02	18.35	11.65	18.43	8.44	9.86	11.98
21.02	17.14	11.42	18.30	8.91	10.79	11.19
03.03	18.27	13.06	20.77	11.95	11.25	14.09
13.03	24.48	17.95	27.47	14.53	15.04	18.77
26.03	26.76	19.33	29.45	16.08	16.61	21.87
06.04.	27.93	18.84	33.55	16.40	16.63	19.87
19.04	27.67	20.20	30.53	16.40	18.57	21.32
04.05	32.36	24.29	38.02	22.09	22.85	26.63
26.05	30.45	24.23	36.14	20.98	22.29	26.04
03.06	32.87	27.27	38.78	22.62	25.00	28.91
03.07	30.75	25.67	33.51	22.24	25.14	26.98
16.07	32.11	27.64	31.33	27.53	23.63	29.05
30.08	40.64	31.04	37.18	29.84	31.44	35.24
22.10	33.00	29.00	31.00	24.00	27.00	32.00
04.12	41.91	37.07	37.15	29.34	36.59	38.46
18.12	43.79	39.64	37.09	29.93	36.26	38.05
Date			Weight in g
2022	1	2	3	4	5	6
Born	04.07	04.07	04.07	04.07	04.07	04.07
Sex	♂	♀	♀	♂	♀	♂
Shedding	20.12	24.12	18.12	13.12	15.12	16.12
2024	10.03	08.03	21.03	15.03	22.03	12.03
16.01	41.60	36.00	36.50	28.30	33.90	38.50
23.02	41.30	35.60	37.90	28.40	34.20	35.80

). In the period July—August 2023, a sharp jump in the growth was noted. With each skin shading, the offspring lag behind in weight gain (

Table 3. Data on the shedding of the juvenile V. a. ammodytes × V. a. transcaucasiana hybrids.

Data			Shedding
2022	1	2	3	4	5	6
born	04.07	04.07	04.07	04.07	04.07	04.07
sex	♂	♀	♀	♂	♀	♂
2022	02.12	24.12	17.12	15.01 23	14.12	03.12
2023	13.02	16.02	14.02	07.03	06.03	13.02
	08.04	28.03	06.04	20.04	18.04	04.04
	17.06	13.06	10.07	17.06	19.06	15.06
	20.12	24.12	18.12	13.12	15.12	16.12
2024	10.03	08.03	21.03	15.03	22.03	12.03

, Figure 3). The juveniles consumed food regularly and only few longer periods of fasting occurred in two of the specimens (Table 1).

A few days after feeding on 23. 02. 24, with a slightly sharper drop in temperature, the juveniles settled in the cool corners of the boxes and prepared for wintering—tightly curled up in the cool end of the box, with the head and tail completely covered by the body. The snakes were irritated by the light, hissed loudly and took defensive position. We gradually lowered the temperature to 12–15 °C, with almost no daylight, and left them to hibernate for about a month. On 01. 04. 24, the preparation for the termination of hibernation began. During hibernation, the juveniles were only observed, without being weighed, to avoid excessive stress as this was their first real “wintering”. All of the juveniles survived the hibernation (Table 1, Table 2 and Table 3).

4. Discussion

In recent years, information has periodically appeared concerning specimens, observed in nature, with characteristics typical for two separate species or subspecies, as well as information about their coexistence and the appearance of intermediate forms [3]. According to some authors, the only factor limiting hybridization are some geographical features of the habitat. However, according to some morphometric and genetic studies, the difference in habitats has no influence on hybridization. In general, hybrid specimens have been described primarily in contact zones [13,14,15]. This suggests that in such zones, individual species and subspecies naturally hybridize with each other, mixing their characteristics in generations and subsequently these signs are assimilated into one of the original parental species or subspecies, if the generation turns out to be fertile [16]. In such a case, hybridization has evolutionary significance in gene transfer (introgression), enhancing heterozygosity, and eliminating the effects of genetic drift [17].

The study of contact zones is essential for understanding hybridization processes. Natural hybridization can be observed in about 10% of species in major faunal groups [18] and is mainly found in closely related taxa due to the absence of reproductive barriers or limited ones, but can sometimes be observed in well-defined species. Reproductive barriers can be pre- or postzygotic [19]. Prezygotic barriers to hybridization include spatial and temporal isolation, natural selection against migrated individuals, sexual isolation, and, after mating, prezygotic isolation. In the absence of prezygotic reproductive barriers, postzygotic barriers can still hinder hybridization, with F1 hybrids often being sterile, while F2 hybrids are often not viable or exhibit reduced fitness. Postzygotic barriers may be associated with either inherent or environmentally induced postzygotic isolation [3,20], while sexual selection against hybrids may constitute another postzygotic barrier [21,22]. If reproductive barriers do not completely isolate two species, introgression may lead to reticulated phylogenies, adaptations, and speciation [18,23]. Introgression does not always affect both species equally, and asymmetric introgression is common [24].

The appearance of hybrids in contact zones may indicates that species and subspecies arose quite late in history and speciation processes are still vigorous and operating at high speed, i.e., the final formation of species and subspecies is not completed and their differentiation occurs only on geographical or ecological basis [4].

Research in the contact zones between V. berus and V. aspis demonstrated that less than 0.1% of the captured specimens were hybrids [2,25], according to a report on a natural hybrid of V. ammodytes and V. berus caught during the summer of 2005 in probably the only known region in Romania, where the two parent species coexist (Metalliferous Mountains, Transylvania). The authors reported that the specimen had severe health issues. According to Czirják et al. [25], in F2 the individual viability is low (see also [24,26]. In our study the hybridization of the subspecies was successful and the hybrids exhibited no pathological traits during their development and were consistent in their growth pattern (Figure 3). The animals exhibited sexual dimorphism characteristic for the genus Vipera. Our data (compared to existing research) might indicate that intermediate subspecies hybrids are probably able to exist in nature, however, it is uncertain if these intermediate forms survive long term or exist as temporary meta-populations or singular individuals. In future research we plan to do a thorough investigation on the hybrids with additional morphological and genetic analysis in order to discover phenotypic and molecular traits which may help researchers detect hybrid viperids in the wild.