The state of Tamaulipas has many multiflora resources that could be used for the development of beekeeping, particularly in the Sierra Madre Oriental (Villegas et al., 2003); in addition, in the center of the state there is an important citrus area (SDR, 2017). However, the africanization of the nuclei has led to the decrease in honey production; also, diseases, parasitosis, lack of training of producers and climatic phenomena have also diminished the productive potential of bees (Guzmán-Novoa et al., 2004; Medina- Flores et al., 2015; Urbina-Romero et al., 2019). Similarly, the introduction of new pests such as the hive beetle (Aethina tumida), recently reported in the country by SAGARPA (2007).

The arrival of Africanized bees in 1987, brought significant changes in the way of beekeeping in the country; these hybrids inherited undesirable characteristics from their African parents (Apis mellifera scutellata), such as defensive behavior, the tendency to swarm and migration (Medina-Flores et al., 2015; Urbina-Romero et al., 2019). The direct consequences of this defensive behavior were the abandonment of the activity of producers, reduction of the number of hives, increase in production costs due to the acquisition of protective equipment and the periodic exchange of fertilized queens (SAGARPA, 2016).

However, Africanized bees have desirable characteristics for apicultural activity, including adaptation to tropical climates that results in rapid population growth and resistance to certain diseases and parasitosis (Vázquez-Castro et al., 2006; Medina -Flores et al., 2015).

On the other hand, the polymerase chain reaction (PCR), which is associated with restriction fragment length polymorphism (RFLP), which are a versatile, fast and low-cost tool to detect single nucleotide polymorphisms (SNPs) of genes associated with production characteristics; as well as to evaluate the genetic variability between and within populations (Sifuentes et al., 2006). It has been proven that polymorphism in nuclear DNA (DNA) and mitochondrial DNA (DNAm) are very useful molecular markers in the study of population genetics (Clarke et al., 2001).

Likewise, the DNA polymorphism was used to discriminate groups of bees belonging to the African and European subspecies (Esquivel-Rojas et al., 2015). On the other hand, Medina-Flores et al., (2015) found in a study conducted in Zacatecas, Mexico, that the differences in the frequencies of African and European morphotypes were significant, both intraregionally and between the three regions.

The present study aimed to characterize Apis mellifera bee populations of commercial apiaries in the state of Tamaulipas, using molecular analysis techniques to identify DNAm variability and the presence of European and African genes.

103 samples were collected from commercial apiaries in the state of Tamaulipas; producers replace their queens approximately every two years (SAGARPA, 2016), with certified queen bees (not Africanized); Some resort to artificial feeding of bees during the cold and dry seasons, with products such as fructose syrup and soy cake.

The samples (bees) were kept in 95 % absolute ethanol alcohol at room temperature until DNA extraction. The complete DNA extraction was carried out in the Molecular Biology Laboratory of the Faculty of Veterinary Medicine and Zootechnics of the Autonomous University of Tamaulipas, according to the protocol proposed by Crozier and Crozier (1993). For the extraction of DNA, four bees were frozen with liquid nitrogen (N2), macerated in sterile mortar and the samples were digested at 55 °C for 16 hours in 295 µL of Master-Mix solution (Proteinase K, RNase A Solution, EDTA 0.5 M, pH 8.0; Nuclei Lysis Solution). To purify the DNA, the 1.5 mL Eppendorf® tube was added to the sample, 250 µL of Wizard® SV Lysys Buffer, centrifuged at 13000 Xg for one minute; was transferred to the collecting tube with a mini-column with filter, 650 µL of Wizard® SV Wash Solution (four times) was added, centrifuging at 13000 Xg for one minute. Subsequently, the minicolumn tube was discarded and the filter was placed with the DNA in a new 1.5 ml Eppendorf® tube, 250 µL of deionized sterile nuclease-free H2O was added by incubating for two minutes at room temperature. The sample was centrifuged at 13000 Xg for two minutes and the DNA obtained was frozen at -20 °C until use.

The PCR was developed according to the protocol described by Esquivel-Rojas et al. (2015), but with 100 ng of DNA, 50 nm of each initiator and 25 µL of Mastermix®, which includes 1 U of Taq polymerase and 10 µL of sterile nuclease-free H2O.

The 485 bp region of the Cytochrome b gene was amplified using the Cytb-b F initiators: TAT GTA CTA CCA TGA GGA CAA ATA TC and Cytb-R ATT ACA CCT CCT AAT TTA TTA GGA AT. The mixture was denatured at 94 °C for one minute, followed by 30 cycles, 94 °C for one minute, 60 °C for one minute and 72 °C, 30 seconds; finally at 72 ºC for 7 min, using a Nyx Technik® Thermocycler. For the PCR-RFLP 50 µg of PCR product was used which was digested with 1.5 U of Bg/II in a final volume of 25 µL. It was incubated for 16 hrs and the digestion products were separated by electrophoresis in a 1 % agarose gel in 1X TBE solution; a standard 50 bp marker was used. The gels were stained with ethidium bromide and the images were captured with the DigiDoc-IT System® photodocumenter and the Launch Doc-ITLS Acquisition® software. A pattern of two fragments (194 and 291 bp) corresponding to the European mythotype (E) and a single undigested fragment of 485 bp corresponding to the African mitotype was visualized (Crozier and Crozier, 1993).

For the detection of the L2S1int polymorphic site, the primers F: GGCGTCCAGGTAACCGTCTCC and R: CGGTTGGAGGCGAACGGAAA were designed to obtain an 830 bp fragment. The PCR conditions are those recommended by Suazo and Hall (2002), using the restriction enzyme AVAI to identify the African allele. Since this fragment analysis were inconclusive to identify the expected molecular characteristics, the results are not included in this manuscript.

For the statistical analysis and to determine the Hardy-Weinberg equilibrium, the heterozygous deficit and the polymorphic content index of the locus used for the characterization of the populations, the Cervus v3.03® and GENEPOP v4.0.10® softwares were used.

The molecular method used in the present work was based on the PCR-RFLP of DNAn and DNAm; On the one hand, the 830 bp L2S1int suballele nuclear fragment was amplified using the restriction enzyme AvaI. Likewise, the mitochondrial fragment of 485 bp of the mitochondrial gene of cytochrome b was amplified using the enzyme Bg/II; since these two loci contain the changes in the sequences that allow discrimination of the African subspecies Apis mellifera scutellata and the non-African subspecies.

Figure 1 shows the agarose gels with the mRNA PCR products, electrophoresed and stained with ethidium bromide (EtBr). The amplified fragment of the cytochrome b gene with a band size of 485 bp (A) and the amplified segment of the suballele L2S1int a fragment of 830 bp (B) was observed. In the present work it was observed that the polymorphism of the mitochondrial gene of Cytochrome b detected with the enzyme Bg/II discriminates the mitochondrial haplotype of A. m. scutellata; ancestor of Africanized bees, from the haplotype of A. m. mellifera, A. m. ligustica, A. m. cárnica and A. m. caucasica (figure 2).

Figura 1 Lanes: 1, 50 bp molecular marker and Lanes 2-8, segment of the amplified cytochrome b gene (A).

The molecular method used in the present work was based on the PCR-RFLP of DNAn and DNAm; on the one hand, the L2S1int suballele nuclear fragment of 830 bp was amplified using the restriction enzyme AvaI. Likewise, the mitochondrial fragment of 485 bp of the mitochondrial gene of cytochrome b was amplified using the enzyme Bg/II; since these two loci contain the changes in the sequences that allow discrimination of the African subspecies Apis mellifera scutellata and the non-African subspecies.

Figure 1 shows the agarose gels with the mRNA PCR products, electrophoresed and stained with ethidium bromide (EtBr). The amplified fragment of the cytochrome b gene with a band size of 485 bp (A) and the amplified segment of the suballele L2S1int a fragment of 830 bp (B) was observed.

In the present work it was observed that the polymorphism of the mitochondrial gene of Cytochrome b detected with the enzyme Bg/II, discriminates the mitochondrial haplotype of A. m. scutellata; ancestor of Africanized bees, from the haplotype of A. m. mellifera, A. m. ligustica, A. m. cárnica and A. m. caucasica (figure 2).

Figura 2 1.7% agarose gel electrophoresis. Fragment of 485 bp characteristic of Africanized bees (lanes 2-4). Segments of 194 Fragment 291 bp characteristic of non-Africanized bees (lanes 5-7).

Electrophoretic analysis of 1.7 % agarose gels with the DNAm of the samples showed fragments of DNAm bands of African and European genotypes (Figure 3). The fragments that maintained the size of 485 bp are characteristic of African bees, (because they lack the cutting site for the enzyme Bg/II in the cytochrome b gene); while in individuals that had the restriction site fragments of 194 and 291 bp were produced, corresponding to non-African bees.

Figura 3 Segments of 485 bp characteristic of Africanized bees (lanes 2-13, 15 and 18). Segments of 194 and 291 bp characteristic of European bees (lanes 14, 16, 17).

It was observed that the populations have a degree of Africanization greater than 55 % (Table 1), with the exception of Jaumave.

Table 1

Locality	N	European (E)	African (A)	E-A
Victoria	15	5(33.3)	9(60.0)	1(6.7)
Altamira	24	2( 8.3)	22(91.7)
Llera	23	5(21.7)	18(78.3)
Güémez	24	8(33.3)	14(58.3)	2(8.3)
Padilla	12	5(41.7)	7(58.3)
Jaumave	5	2(40.0)	2(40.0)	1(20.0)
Total	103	27(26.2)	72(69.9)	4(3.9)

On the other hand, genotypic and allelic frequencies in bee populations in the Tamaulipas regions are presented in Table 2.

Table 2

Population	N	Genotypic Frequencies
		European (E)	African (A)	E/A
Victoria	15	0.333	0.600	0.067
Altamira	24	0.083	0.917
Llera	23	0.217	0.780
Güémez	24	0.333	0.583	0.083
Padilla	12	0.417	0.583
Jaumave	5	0.400	0.400	0.100

E/A = European / African

The performed analyzes for the genetic frequencies of the populations indicated that they are not in Hardy-Weinberg equilibrium, probably because there is a heterozygous deficit.

The molecular method used allowed discrimination of the African subspecies Apis mellifera scutellata and non-African subspecies (Zamora et al., 2008 Esquivel-Rojas et al., 2015). The polymorphism of the mitochondrial gene of cytochrome b detected with the Bg/II enzyme discriminates the mitochondrial haplotype of A. m. scutellata; ancestor of Africanized bees, from the haplotype of A. m. mellifera, A. m. ligustica, A. m. cárnica and A. m. caucasica (Pinto et al., 2003 Genchi-García et al., 2018).

Electrophoretic analysis of 1.7 % agarose gels with the DNAm of the samples showed fragments of DNA bands of African and European genotypes. The fragments that maintained the size of 485 bp are characteristic of African bees, (because they lack the cutting site for the Bg/II enzyme in the cytochrome b gene); while in individuals who had the restriction site, fragments of 194 and 291 bp were produced, corresponding to non- African bees.

The pattern observed in this work coincides with the pattern to discriminate African haplotypes from non-Africans in Africanized areas (Pinto et al., 2003 Tibatá et al., 2018). It was observed that the populations have a degree of Africanization; similar results were reported by Esquivel-Rojas et al. (2015), Medina-Flores et al. (2015), Quezada-Euán (2007) and Zamora et al. (2008), when evaluating the polymorphism of Africanized populations in Mexico.

The results obtained with the modifications to the DNAm purification protocol, allowed to know the genetic structure of honey bees in the center of Tamaulipas. The characterization of the populations showed that 69.9 % of the colonies evaluated had in their DNAm, the restriction site for the African haplotype; as noted in the literature (Quezada-Euán, 2007; Zamora et al., 2008 Esquivel-Rojas et al., 2015; Medina-Flores et al., 2015). While 26.2 % of the sampled apiaries had the haplotype in their DNAm, which characterizes them as of European maternal origin. 4 % of the samples presented mixed European-African haplotype; this could be due to drones being wild and presenting a high proportion of African genes. However, Tamaulipas beekeepers have as strategies to introduce queens every year or year and a half, but certified before the Ministry of Agriculture, Livestock, Rural Development, Fisheries and Food (SAGARPA), if they are not carriers of the African gene (SAGARPA, 2016 ).

In the sampled sites, the African haplotype represented a high percentage of the colonies, confirming the reduction of European type alleles in the beekeeping populations of Mexico, Colombia and Brazil (Quezada-Euán, 2007); and the dominance of African nuclear and mitochondrial genes in the genome of domestic colonies, from the arrival of the Africanized bee in Mexico.

Altamira populations are composed almost entirely of bees with mitotypes of African origin, probably because the producers do not make the annual change with queens of European origin. These results suggest that beekeeping practices, such as the annual change of queens, have not been sufficient to reverse the frequency of African genes in local populations. In addition, the tropical climatic conditions of the region have facilitated the displacement of European genes.

It was observed that populations have a high degree of Africanization, with the exception of Jaumave; which indicates that it is still necessary to continue introducing European certified queens (SAGARPA, 2016).

Quezada-Euán (2007) reported 95.0% of Africanization for Chiapas and Tabasco; Uribe et al. (2003) and Zamora et al. (2008) reported 13.7, 48.0, 50.0 and 21.0 %, of haplotypes derived from African bees in Sonora, Baja California Sur and Baja California Norte, respectively. Similarly, Quezada-Euán (2007) reported that the proportion of Africanization in Michoacán and Jalisco was 56.0 and 40.0 %, respectively. While in Yucatan, it increased from 61.0 to 73.0 %. Alaniz-Gutiérrez et al. (2016) observed that African morphotypes in Ensenada and Mexicali, Baja California are above 85.0 %.

It can be concluded that the populations of Apis mellifera L. de Tamaulipas, are constituted by characteristic genes A. mellifera scutellata and to a lesser extent by genes of European subspecies such as A. mellifera ligustica. The results of the molecular marker analysis showed that Tamaulipas populations are heterogeneous with introgression of African genes in European populations.