Heat stress is a condition in which animals become unable to dissipate heat load generated from their body metabolism and environment, thereby failing to maintain their body thermal balance (1–3). Heat stress leads to an increase in rectal temperature (RT), respiration rate (RR), core-body temperature (CBT), panting score (PS), pulse rate (PR), sweating rates (SR), and heart rate (HR) (4). Heat stress has adverse effects on milk production traits, fertility and heath in dairy cattle (5), especially in tropical countries where the climate is always hot and humid (6). Dairy cattle in these regions are exposed to higher ambient temperature (Ta), relative humidity (RH), solar radiation (7) and wind variations for extended periods of time. In dairy cattle, temperature and humidity, usually combined as temperature-humidity index, (THI) surpassing their thermoneutral zone cause high reduction in dry matter intake, milk yield, and fertility, which are associated with physiological changes including CBT, RR, PR, RT, and SR (8, 9). These effects on milk production and composition, including fat, protein, lactose, and solids-not-fat percentages, can directly impact the income generated from dairy farming (10). It is crucial for dairy cattle to be within its thermoneutral zone, where energy expenditure to maintain internal body temperature regulation is not required (3). The thermoneutral zone for Bos taurus dairy cattle breeds typically ranges from 0.5°C to 20°C (8), although this can vary depending on factors such as production status, acclimatization level, pregnancy status, feed type, and climatic conditions including wind speed, solar radiation, and relative humidity (3, 8).

Even in countries with moderate temperature such as in the European Union (EU) and the United States of America (USA) dairy cattle can still experience heat stress (HS) due to the effect of climate on evaporative cooling (3). Heat stress has been shown to cause significant economic losses in the dairy industry, with annual losses ranging from 1.69 to 2.26 billion dollars in the USA and a reduction of 0.73 million liters of milk in India in 2020. In the United Kingdom (UK), where the majority of dairy herds are composed of Holstein-Friesian breeds which are known for high milk yield, the projected annual average milk yield loss is 2.4% (~170 kg/cow) in the South West regions (3). Genetic selection for heat tolerance in dairy cattle without compromising their milk yield potential is seen as a sustainable strategy to support dairy cattle production in tropical environments. Dairy cattle breeds in tropical countries are diverse, including heat-adapted Bos indicus cattle and less resilient Bos taurus breeds, and they may respond differently to the effects of HS due to their genetic makeup (11). Therefore, genetic evaluation for thermotolerance and selection of heat-tolerant animals based on specific regions and climates is possible and is a cost-effective measure to improve dairy cattle production in hot weather conditions. It should be noted that milk production traits are often antagonistic to heat tolerance, and genetic selection for higher milk production without considering heat tolerance may result in increased susceptibility to HS (4, 12, 13). Therefore, a balanced approach that considers both milk production traits and heat tolerance is necessary for effective genetic selection in dairy cattle breeding programs. The study by Dikmen et al. (14) reported that heat-tolerant Holstein cows had higher milk yield and better reproductive performance compared to heat-susceptible cows in a tropical environment. Sigdel et al. (15) found that crossbreeding Holstein cows with heat-tolerant breeds such as Sahiwal and Gir improved milk yield and reproductive performance under HS conditions. Furthermore, Ekine-Dzivenu et al. (13) demonstrated that incorporating thermotolerance as a selection criterion in breeding programs can improve milk production.

In addition to genetic selection, environmental modifications can also be implemented to mitigate the impact of HS on dairy cattle. These may include providing shade, improving ventilation, using evaporative cooling systems, and adjusting feeding and watering practices. However, these measures may not always be cost-effective or feasible, especially in resource-limited tropical environments (4, 6). Therefore, integrating genetic selection for heat tolerance into breeding programs can be a more sustainable and long-term solution to mitigate the adverse effects of HS on milk production traits, fertility, and health in dairy cattle. Therefore, the objective of this review was (1) to characterize the genetic and phenotypic responses to HS, (2) identify candidate genes responsible for heat tolerance; and (3) discuss mitigation strategies for dairy cattle breeds reared in tropical countries. The findings of this review will contribute to policy formulations with regard to strategies for preventing the effects of HS on dairy cattle breeds in sub-Saharan Africa, where the climate is very hot and humid almost year-round.

Cattle breeds used for milk production in Africa are categorized into four groups: the humpless Bos taurus, found mostly in west Africa; the humped Bos indicus (Zebu), temperate taurine, which have been disseminated extensively in Africa; Sanga which is a crossbred of B. taurus x B. indicus, predominantly reared in eastern and southern Africa; and Sanga x Zebu cross well-known as Zenga (16, 17). The dominant cattle breed reared in Africa is the Zebu cattle, which is effectively adapted to the tropical climates and tolerant to extreme temperatures and HS. Moreover, the breed is highly resistant to diseases and parasites like ticks and has low nutritional requirements (18). However, the breed has low potential for milk production.

In Sub-Saharan Africa (SSA), breeds commonly used for dairy production are mainly crossbreds from local breeds and imported animals or semen from USA, Canada, Europe, Australia and New Zealand. Holstein is the preferred exotic breed for milk production in SSA, followed by Jersey, Ayrshire and Guernsey (19, 20). Among the dairy cattle breeds introduced in Africa, the Holstein dairy cattle breed is the predominant breed in most countries and it is the more productive genotype worldwide (21). Additionally, the Holstein genotype is more favored by farmers due to its ability for high milk production, which lead to high profits at the farm. Jersey, Ayrshire and Guernsey genotypes are considered as lower milk producers and are, then, less preferred. On the other hand, Jerseys are preferred owing to the high butterfat content of their milk and the relatively lower rearing costs due to their small size (20). It is well known that the Holstein and Jersey breeds differ in their abilities to respond to HS effects. Many studies have indicated that the Jersey breed is more tolerant to HS compared to Holstein dairy cattle based on milk production decline (19).

Because of lack of adaptability of pure dairy cattle breeds to tropical environment, several countries in sub-Saharan Africa have resorted to using crosses of Holstein-Friesian, Ayrshire, Jersey, with Bos indicus for milk production. Crossbreeding of indigenous cattle with exotic dairy breeds has been utilized to improve milk yield in hot regions and attaining adequate reproduction rates. Currently, there are a large number of crossbreds produced by crossing the exotic dairy genotypes and local Zebu cattle which are used as dairy cows (19) in sub-Saharan countries. In most tropical countries crossbreeding has been used to combine high production traits of B. taurus (Holstein) with greater heat tolerance of B. indicus (Zebu). In some countries, this strategy has led to development of a higher producing and heat-tolerant genotype such as Girolando genotype in Brazil (22). In Thailand, crossbreeding between Bos taurus (Holstein, Jersey, and other dairy cattle breeds) and Bos indicus (Sahiwal, Red Sindhi, and Thai Native) is predominant and the main crossbred dairy cows (>87.5% Holstein genetic composition) are crosses of Holstein and Sahiwal or Thai Native breeds (23). Studies in tropical countries have indicated that Holstein-Friesian crossbreds have relatively better milk production performance while Jersey crossbreeds have higher fertility in comparison to the anterior (24). In addition to crossbreds, Bos indicus genotypes, e.g., Sahiwal and Gir, which are regarded to be highly resilient to tropical environment and slightly sensitive to HS are used for milk production (25). Table 1 shows different dairy cattle breeds reared in tropical countries for milk production.

The indicators of the impact of HS in animals include: (1) physiological parameters (e.g., respiration rate, rectal temperature, core-body temperature, sweating rate, panting score, among others); (2) milk production traits; (3) oxidative stress biomarkers; and (4) genes influencing HS tolerance. Additionally, the THI which combines the effects of ambient temperature and relative humidity, wind speed and solar radiation is also a proxy for HS conditions, and may approximate potential HS in dairy cattle. A comprehensive review about those HS biomarkers was presented in our previous review (45). Table 2 shows indicators of HS in dairy cattle and methods used to measure them.

Dairy cattle have developed three coping mechanisms, i.e., acclimation, acclimatization, and adaptation to reduce the impacts of HS on their biological systems (59, 60). Acclimation is the coordinated phenotypic response developed by dairy cattle to a particular environmental stressor while acclimatization is a coordinated response to numerous simultaneous stressors such as temperature, humidity, wind speed, and solar radiation (59, 60). Both acclimation and acclimatization act to improve dairy cattle fitness to the changing climatic conditions (59). On the other hand, adaptation involves genetic changes as negative effects of environmental variables on an animal persist over long time (59, 60). To cope with HS conditions, dairy cattle spend long time standing and reduce activity to rise surface area for heat abatement, sensible water loss, radiating surface area, and air movement via convection, conduction, radiation, and evaporation (61, 62). Heat stress prevents dairy cattle to fully express their genetic potentials for milk yield and affects milk composition (63). Dairy cattle react to HS through: (1) decline in feed intake, (2) high water consumption, (3) altered basal metabolic rate, (4) increased evaporation of water, (5) physiological changes, i.e., higher body temperature, higher respiratory rate, panting rate, pulse rate, heart rate, rectal temperature, and (6) altered blood hormone concentration (45). A dairy cow experiencing HS produces small quantity of milk and has a high risk of disease infection (61). Under HS conditions, the decrease in feed intake has been recognized as an important cause of decline in milk production as it is associated with a negative energy balance state regardless of the lactation stage (41). Feed intake in lactating Holstein cows begins to decline at the ambient temperatures of 25–26°C and decreases sharply when the temperature is over 30°C. At 40°C, feed intake can decrease by 40% (41).

Heat stress is associated with reduced milk, fat, and protein yields (64–66). For instance, Negri et al. (67) reported 21% milk yield losses in a commercial herd of Holstein dairy cattle reared in southern part of Brazil due to HS. In Bangladesh, Chanda et al. (41) indicated that when a lactating Holstein cow is moved from an ambient temperature of 18 to 30°C milk yield is decreased by around 15%, followed by a 35% decline in the efficacious of energy usage for production reasons.

The THI is used to quantify environmental heat load on animals. Milk yield is negatively associated with THI. In Thailand, Sungkhapreecha et al. (23) found a smaller decrease in milk production of about 70–80 g per point increase in THI value of 76 for cows with >93.7% Holstein genetic composition, and a 20–40% lower milk production in comparison to other populations. In this study, crossbred dairy cattle >87.5% Holstein genetics were crossbreeds between Holstein and Sahiwal or Thai Native breeds (23). Moreover, milk yield, milk fat, solids-not fat, and protein percentages are reduced by 15, 39.7, 18.9, and 16.9%, respectively, when a lactating Holstein cow is transferred from an ambient temperature of 18 to 30°C (68). Table 3 presents information about dairy cattle phenotypic responses to HS for studies conducted in tropical countries of Asia, Africa and Latin America.

Heat stress induces modification of metabolites in the mammary glands of lactating dairy cattle, along with glycolysis, lactose, ketone, tricarboxylic acid cycle (TCA), amino acid, or nucleotide metabolism, which prevents the supply of substances for milk production in the mammary gland of lactating Holstein cows (82). Therefore, HS alters milk synthesis and composition by altering metabolisms of substances in the mammary gland tissues of lactating dairy cows.

Milk and blood metabolites are used to provide accurate information on the degree of HS in dairy cows as they change during HS (83). They also help in the genetic selection of genotypes and individual animals that are heat tolerant (82). These metabolites are identified in animal biofluids (e.g., serum, plasma, urine, and milk) through metabolomics approaches. A metabolomics approach is considered as potential method for extensive metabolite analysis of milk samples. Three analytical approaches are extensively used to study the milk metabolome: nuclear magnetic resonance spectroscopy (NMR), gas and liquid chromatography-mass spectrometry (GC–MS/LC–MS). Among these approaches, NMR can provide information on the direct association between metabolite richness and resonances.

LC–MS has greater sensitivity and reproducibility, together with a superior dynamic scale, becoming more worthy for the investigation and determination of non-volatile and greater macromolecular metabolites (84). Untargeted metabolomics applying LC–MS is a reliable method for the identification of a myriad of metabolite changes in a biological matrix, like blood, urine, cells, and tissues (85). On the other hand, proton nuclear magnetic resonance (NMR) combined with multivariate statistical analysis are very useful for metabolite profiling to investigate the metabolic variations in blood, milk and liver of heat stressed dairy cattle (86). Blood and milk metabolites (e.g., glucose, cholesterol, non-esterified fatty acids, blood urea nitrogen, creatinine, among others) which are identified through metabolomics are generally used to investigate the degree of HS in dairy cattle (51, 87). In the study conducted by Zheng et al. (54) at Heilongjiang province in China in Holstein dairy cows, seven blood metabolites including L-leucine, L-isoleucine, butanoic acid, and L-valine were significantly negatively associated with reactive oxygen species (ROS) while four milk metabolites, i.e., galactinol, undecane, cis-aconitate, ethanol were significantly positively correlated with ROS. The above metabolites are involved in the metabolic pathways associated with amino acid, carbohydrate, translation metabolism and metabolism of cofactors and vitamins. Both blood and milk metabolites were identified using GC–MS.

Using LC–MS and ¹H-NMR, Tian et al. (88) in China, identified 53 potential diagnostic biomarkers for HS from milk samples of heat-stressed lactating Holstein dairy cattle. Included among them were lactate, pyruvate, citrate, creatine, proline, β-hydroxybutyrate, among others. The above metabolites are involved in various biochemical pathways including carbohydrate, amino acid, lipid and gut-microbiota-derived metabolism. Yue et al. (83) investigated the HS effects on milk and blood metabolites using proton nuclear magnetic resonance (¹H-NMR) between HS groups and non-HS lactating Holstein dairy cattle in China and identified eight metabolites from milk samples and twelve metabolites from plasma samples which were significantly unrelated between the heat stressed and non-heat stressed-experimental groups. The identified metabolites include alanine, glycose, glutamate, urea, histidine, leucine, lipid, β-hydroxybutyrate, citrate, choline, formate, pyridoxamine, among others. The above metabolites are involved in different metabolic pathways including proteolysis, glycolysis, gluconeogenesis, lipolysis, galactose, purine and alanine metabolism, among others. Hu et al. (89) investigated the effects of HS on plasma metabolites between Inner-Mongolia Sanhe and Holstein cattle in China using ¹H-NMR combined with multivariate statistical analyses. In their study, 26 potential diagnostic biomarkers for HS (valine, leucine, isoleucine, glutamate, glutamine, creatinine, among others) involved in amino acid, carbohydrate and lipid metabolic pathways were identified. Fan et al. (82) identified 33 potential metabolite biomarkers that can be used to identify the degree of HS in lactating Holstein dairy cattle in China. Among them glucose, lactate, β-hydroxybutyrate, creatinine, fumaric acid, and glycine were detected as indicative biomarkers for the degree of HS in lactating dairy cattle using LC–MS/MS. The afore-mentioned metabolites are involved in glycolysis, ketone, TCA, amino acid and nucleotide metabolic pathway. In this study, HS-persuaded changes in glycolysis-associated metabolites were detected for glucose, lactate, pyruvate, fructose 1,6-bisphosphate, among others (82).

In the subsequent study by Fan et al. (85), also conducted in China on Holstein dairy cattle, HS altered energy metabolism, including the decline in the concentrations of glucose, lactose, and galactose-1-phosphate, the increase in the concentrations of acetoacetate and β-hydroxybutyrate (ketone metabolites), and the alterations of TCA-associated metabolites (85). In their study, they identified 34 milk metabolites which are involved in amino acid, glycolysis, lactose, ketone, TCA, and nucleotide metabolic pathways (85). The alterations in nutrient splitting may impact the nutrient usage and the milk production in mammary gland. Findings from these studies indicate that identification of diagnostic biomarkers for HS can contribute to a better understanding of the biological mechanisms of responses to HS (89). This is very important for genomic selection of heat-adaptive dairy cattle and breeding programs aiming at improving heat tolerance in lactating dairy cows (82). Moreover, analysis of these metabolites can help to interpret the physiological mechanisms of HS-persuaded metabolic diseases and lactation (85).

On the other hand, HS alters blood parameters such as packed cell volume, white blood cells, red blood cells, hemoglobin, glucose, electrolyte concentrations, leukocytes, red cell distribution size, mean corpuscular volume, and platelets, among others (22, 72). For instance, neutrophilia happens in heat-stressed cattle owing to the collection of neutrophils from the bone marrow toward the portal blood (22). Red blood cell width (RDW) is not affected by elevation of THI and physiological parameters, implying that heat does not alter cell configuration and increased THI does not influence mean cellular hemoglobin concentration (MCHC) levels (22). Mean platelet volume (MPV) has been found to increase with THI and physiological traits (besides heart rate and panting score) (22). Kekana et al. (90) investigated the effects of HS on pregnant Jersey dairy cattle reared in Limpopo province of South Africa and found that at THI of 75–87, all lactating dairy cattle exhibited higher total protein, blood urea nitrogen and creatinine at 21^st and 75^th days of milk production. In the study by Jeelani et al. (72) conducted in the sub-tropical region of India on 33 crossbred dairy cattle of Bos taurus (Holstein Friesian) x Bos indicus (Hariana/Tharparkar), both red blood cells (RBC) and white blood cells (WBC) counts were significantly unrelated from those determined at THI of 72–74. Nevertheless, at THI of 76, the WBC counts were elevated while RBC counts reduced. The general trend showed that the rise in THI is correlated with elevation in WBC counts and decline in RBC counts. Gaafar et al. (52) investigated the effects of HS on blood parameters of Friesian suckling calves during winter and summer season in Egypt and found a significant effects of THI on different blood parameters. Generally, average trends of hemoglobin (HGB) concentration and count of white blood cells (WBC) and red blood cells (RBC) decreased significantly in hot season than in cool season as a result of HS.

Moreover, hematocrit percentage (HCT) and mean cellular volume (MCV) among other blood parameters increased significantly in hot season than in cool season as a result of HS (22). The decrease in WBC counts observed in the summer season might result from the body system’s response to stress stimuli after coming from the long winter season. In the study by Gao et al. (91) conducted on lactating Holstein cows in China, HS tended to decrease plasma glucose (8%) and non-esterified fatty acids (NEFA) (39.8%), and tended to increase blood urea nitrogen (BUN) (14.7%). Heat stressed dairy cattle showed reduced total plasma amino acids (17.1%), including some essential amino acids (Ile 42.5%, Lys 18.2%) and non-essential amino acids (Ser 10.7%, Arg 25.1%, Gly 20.8%, Cys 27%) (91). Rewe et al. (28) showed that under HS conditions, lactating dairy cattle exposed to increased temperature show elevated blood parameters including erythrocytes count, hemoglobin total concentration, and hematocrit, among others together with urinary pH and density, and the fecal dry matter content.

The quantification of genetic effect relies on the estimates of additive genetic variance and genetic relationships between the traits in dairy cattle populations (38). Estimation of genetic parameters for heat tolerance are crucial for designing optimum breeding schemes and for predicting selection response. To characterize this relationship, the threshold of comfort and the production decline after a particular THI threshold have been used through different genetic models (92). Several studies have determined the genetic effect of HS on milk production traits and estimated the genetic components of HS in dairy cattle (47, 92–96). The statistical models used in the genetic evaluation studies applying test-day milk production records include (1) repeatability models which consider equal genetic relationship between every test-day milk records, (2) single trait and multiple traits models applying every test-day milk record as different parameters, and (3) random regression models (RRM) which treat a covariate function of repeated test-day milk records overtime (97). A more detailed description about these models was made in our previous review (45). The RRM have the potential to analyze every test-day milk records assuming that genetic and non-genetic variances change with days in milk (DIM), parity, lactation stage as do genetic and non-genetic relationships (97). The RRM also accommodate the additive genetic and permanent environmental variance within lactation (93).

Random regression models allow the estimation of genetic variance components and breeding values along the entire trajectory of a time-dependent such as days in milk (DIM) or environment-dependent descriptor like THI covariate (98). In RRM, every animal has two genetic effects, a steady effect that corresponds to animal performance in thermoneutral zone, and a HS effect corresponding to the rate of milk production decay under HS conditions (93). Modeling the effect of a breed as a function of time and environment allows the identification of GxE interactions via differences in genetic covariance components on various combinations of DIM over THI (46). On the other hand, optimization of genetic models that describe the genetic effects of HS on dairy cattle in tropical countries has been challenging because of a lack of proper animal recording. Traditionally, the broken line (BL) model has been used to describe the dairy cattle productive response to increased heat loads (47). This model assumes that production does not change within dairy cattle thermoneutral zone where no response to increasing temperature is observed and that after a particular breakpoint which indicates the beginning of HS, production declines linearly (47, 99). Since animal patterns of response to HS differ depending on climatic conditions, breed type (e.g., Bos taurus versus Bos indicus), production systems, lactation stage (lactating versus non-lactating), and physiological status of the animal, an alternative RRM fitting reaction norms and Legendre polynomial functions is considered rather than the BL model (45, 99).

Although interpreting RRM parameters biologically can be difficult, they are very useful for two reasons. First, they have higher flexibility to fit a regular pattern of decline and, second, they allow the application of eigen decomposition of the additive genetic covariance matrix to detect the selection criteria (92). Such criteria of selection are not associated between themselves and might support the genetic improvement of heat tolerance without having effect on degree of production (92). Moreover, in test day models used for selection on milk production traits, one of the eigen functions has been correlated with the persistence of lactations. This eigen function has been also recommended as a criterion for selection for this parameter to avoid the problems of antagonistic correlations between degree of production and other persistency measures (92). The above two approaches were named the splines (SP) model and the random regression reaction norm (RN) model based on Legendre polynomial (LP) functions, and have been applied to estimate the genetic parameters for heat tolerance in dairy cattle breeds reared in tropical countries. Negri et al. (79) applied RRM with LP functions of 3^rd order to milk production traits and included a model of THI in Brazil, and found a considerable genetic variance for heat tolerance. Table 4 summarizes information about studies that explored the genetic models and genetic parameters for heat tolerance in dairy cattle breeds reared in tropical countries.

Boonkum et al. (6) estimated the genetic effects of HS on milk yield traits using repeatability test models for Holstein crossbreds in Thailand. In this study, they observed a THI threshold ranging from 74 to 82. Sungkhapreecha et al. (5) used single-step genomic restricted maximum likelihood (ssGREML) and repeatability model and observed THI thresholds of 72–82. In a subsequent study, Sungkhapreecha et al. (23) used single-step genomic best linear unbiased prediction (ssGBLUP) and observed a THI threshold of 76. In this study, the genetic correlations between additive genetic effects were negative for all the traits using both methods. However, the positive additive genetic effects were observed for SCS for both methods.

In Brazil, Santana et al. (43) estimated the genetic parameters for milk yield traits using RRM and observed a HS threshold value of 69 with a MY loss estimated at −0.019 kg/year. In this study, it was observed that Gir dairy cattle population were becoming more susceptible to HS. They also found that smaller responses to selection are expected for combinations of THI and DIM values. In subsequent studies, Santana et al. (77) reported THI thresholds ranging from 54 to 79 and a genetic antagonism between milk production performance and HS resilience. This is an important aspect for tropical countries with a hot climate that aim to genetically select for higher milk production but also animals which are highly resistant to extreme temperature and humidity. Santana et al. (78) also used two-trait random regression animal models to estimate genetic parameters for MY, fat percentage (F%), protein percentage (P%), and SCS and observed a THI value of 76 as HS threshold. In this study, 53% of TD milk records were obtained at THI values above 66, and other 14% at THI values greater than 72. There was a decline in milk yield estimated at 0.5 kg/day per cow per THI. They also observed a reduction in MY, SCS, F % and P% at THI values ranging between 66 and 72. They also noted that response to selection can vary as a function of THI and DIM. In another study, Santana et al. (44) estimated the genetic parameters for MY records of Guzera cattle using RRM. In this study, they observed THI thresholds ranging from 72 to 90. Negri et al. (79) estimated the genetic components of MY for Brazilian Holstein cattle using RRM which fitted Legendre orthogonal polynomials, linear splines and Wilmink function. In this study, they obtained a THI threshold of 84 as HS threshold with an average milk yield/day of 23.95 ± 1.9 kg/per THI per cow for THI˃74. They observed that environmental variation increased as the additive genetic variation declined, this directly interfered the h² estimates of MY trait.

In a subsequent study, Negri et al. (67) obtained a THI threshold of 74, associated with MY losses estimated at −0.106 kg/day/THI per cow. Diurnal temperature variation (DTV = 13) was associated with −0.45 kg/day of MY losses. They recommended that it is necessary to include the climatic variables in the genetic evaluation models for genetic selection of high performing sires as many sires could be penalized owing to their daughters being affected by HS. Dauria et al. (4) estimated the genetic parameters of milk fatty acids of Brazilian Holstein dairy cattle using single-trait repeatability test-day models and RRM. In this study, they observed a THI threshold of 68. Carrara et al. (100) estimated the genetic parameters for MY traits and fatty acids of Brazilian Holstein dairy cattle using single trait animal model and RRM. In this study, they obtained a THI threshold of 60. Also, they observed a small variation in the variance components over THI, indicating moderate to low HS for dairy cattle as a result of mitigation strategies using fans, sprinklers, shade trees, and other cooling devices applied at the farm. Otto et al. (101) estimated the genetic components of RT for Gir x Holstein crossbred dairy cattle using GWAS. In this study, they obtained a THI threshold of 97 (42°C and 60% RH). In Tanzania, Ekine-Dzivenu et al. (13) estimated the genetic components of MY for mixed breed dairy cattle using fixed regression model with RRM on a function of THI. In this study, they identified THI values ranging between 67–76. They found that genetic selection for higher milk production disregarding heat tolerance will lead to a deterioration of heat tolerance in this dairy cattle population. In conclusion, RRM fitting RN and LP functions are more suitable than other stated models during the genetic evaluation for milk production and composition traits over different THI thresholds in heat tolerance studies. The RN functions are more convenient to continuous environmental descriptors like THI as they help in the distinction allying individuals that are more or less affected by environmental variations (100). Additionally, through RN functions, it is possible to obtain two genetic components for each individual, i.e., general production capacity and specific capacity to respond to HS (77).

Understanding the mechanisms of climatic adaptation and heat tolerance of dairy cattle breeds reared in tropical countries is vital for the optimization of breeding schemes and conservation of genetic resources (102). Genetic variation in heat tolerance at physiological and cellular levels between Bos indicus and Bos taurus breeds reared in tropical countries has been estimated in various studies (103). Heat tolerance can be identified based on the expression or activation of particular biological biomarkers including heat shock transcription factor (HSF), heat shock proteins (HSP70, HSP90, and HSP27), and Slick Hair gene (36). HSF and HSPs known as molecular chaperons, are crucial genes associated with heat tolerance (37). They play an important cytoprotective role during cellular stress by avoiding the formation of non-functional proteins, maintaining protein folding, and preventing protein aggregation which result into protein homeostasis during cellular response to HS (37, 104). HSPs also activate the immune response mechanisms during HS (37). Heat stress causes misfolding of proteins at cellular level and a cell fails to perform DNA replication, transcription, polypeptide synthesis, and transport of materials (105). Given those HS effects, HSPs are activated and reduce accumulation of the denatured proteins in the cell and enhance its survivability and ability to overcome oxidative stress and thermal stress (105). Moreover, HSP70 act as molecular chaperons and play crucial roles in cellular thermotolerance, apoptosis, immune-modulation and heat tolerance (37, 105). HSP70 function as inducible chaperons and provide a second window of protection while stabilizing the native conformation of proteins and maintaining cell survivability during HS conditions (37, 106). An early induction of HSP70 is associated with the development of heat tolerance in dairy cattle (37). Moreover, HSP40 which harbors DNAJA1, DNAJB1, and DNAJA2 genes, regulates the ATPase activity of HSP70 through interaction with the J domain of the HSP70 proteins (104). Additionally, HSP40 and HSP90 function as co-chaperons of HSP70 and play important roles in the restoring of protein conformation after heat shock (106).

Studies have indicated that worldwide, there exists five continental groups of cattle, i.e., European taurine, Eurasian taurine, East Asian taurine, Chinese indicine and Indian indicine (107). Heat tolerance, ectoparasite resistance, low maintenance cost, and low susceptibility to the gastro-intestinal parasites are important physiological traits of Indicine over taurine (108, 109). The above physiological traits help Bos indicus breeds to grow under tropical and subtropical climates of Africa, South-east Asia, Brazil, Northern Australia, Southern China and parts of the USA (107, 108). The taurine breeds which have increased metabolic rate and feeds requirements are mainly reared in Europe, Northern America, and Australia (108).

To better understand the variation in HS response between Holstein-Friesian (taurine) and Sahiwal (indicine) in India, Kishore et al. (106) investigated the temporal profile of HSP40, HSP60, HSP70, and HSP90 candidate genes using qRT-PCR at 42°C. In their study, they observed that Holstein-Friesian was the most affected by the heat shock than Sahiwal, indicating that the latter is highly adaptable to HS conditions. In another study conducted in Brazil, Otto et al. (101) used GWAS through breed of origin alleles (BOA) to compare the HS response between Holstein and Gir dairy breeds. In their study, it was observed that the proportion of animals with HH and GG alleles were 2.9 and 49.9%, respectively for higher RT in Holstein and Gir, indicating that Holstein dairy cattle were more affected by HS than Gir animals (101).

In Nigeria, Morenikeji et al. (110) used RNA transcriptomic approach to compare HS response between indicine (white fulani and N’Dama) and taurine (Angus and Holstein) dairy cattle breeds and identified six genes (ASIP, MC1R, TYRP1, TYRP2, TYR, and ALOX12E) to be responsible for heat tolerance. These six genes were upregulated in indicine than in taurine breeds. In particular ALOX12E gene was found to be associated with water retention in indicine to avoid dehydration, an adaptive trait important for white fulani and N’Dama survivability in tropical climates (110). Furthermore, Onasanya et al. (111) indicated that HSP70 and HSP90 genes identified in West Africa zebu breeds are thermo-regulatory candidate genes which are less sensitive to thermal signals such as high temperatures. These heat shock protein genes are predominant in the west and east Africa zebu such as white Fulani, N’Dama, Sahiwal, Boran, and Ankole which constitute more than 90% of cattle population in Africa (111).

Selection signature analyses have the potential to identify specific genomic regions and candidate genes associated with adaptive or productive traits of different dairy cattle breeds reared in HS environments of tropical countries (40, 112, 113). Using selection signatures and copy number variations to differentiate Bos taurus and Bos indicus breeds, Jang et al. (114) identified 313 candidate genes under selection and adaptation. A higher copy number of HMGA2 gene was identified in Bos indicus than in Bos taurus breeds. They also identified DNAJC18 gene in indicine specific deletion than in taurine. The DNAJC18 candidate gene has been shown to play a significant role in tropical adaptation and heat tolerance of East African short horn zebu (114). At China-Myanmar border, Sun et al. (107) conducted a study on Lincang humped cattle and identified MED16, DNAJC8, HSPA4, FILIP1L, HELB, BCL2L1, and TPX2 as candidate genes responsible for heat tolerance and parasite resistance. In India, Saravanan et al. (36) studied the selection signatures between taurine breeds and indicine breeds and identified candidate genes responsible for immune response and heat tolerance. Among those genes identified include NCR3 (in Gir), ARID5A (in Ongole), HIST1H2BN (in Sahiwal), BEFB4 and DEFB7 (in Tharparkar) for immune response. Genes responsible for heat tolerance included HSPA1L and HSPA1B in Gir, DNAJB4 in Ongole and GRXCR1 in Tharparkar (36).

Genes such as OLA1, SP9, HSPA12A were associated with heat tolerance in all Indicine cattle breeds (Gir, Hariana, Kankrej, Ongole, Red Sindhi, Sahiwal, and Tharparkar) but none of these genes were identified in taurine breeds (36). The majority of candidate genes identified in taurine breeds were all related to milk production traits and included CACNA2D1, ADARB2, WDR37 and LARP4B (Ayshire), KCNIP4, SLIT2, ADAP2, and LTF (Brown Suiss), SYN3, TIMP3, RFK, ACACA, HNF1B, and SCAP (Guernsey), ADARB2, WDR37, LARP4B, ZMYND11, PRNP, and CDH4 in Holstein Friesian, and LPHN2, ADARB2, WDR37, and LARP4B in Jersey (36). The genes which were found to be associated with heat tolerance in Indicine cattle breeds can be used for molecular detection of dairy cattle breeds with potentialities for heat tolerance. Additionally, the candidate genes identified in various studies can serve as the basis to explore the use of indicine breeds in crossbreeding schemes in order to transfer their adaptation abilities and rusticity to taurine breeds (115). Table 5 provides examples of candidate genes responsible for heat tolerance that have been identified in different dairy cattle breeds reared in tropical countries.

Heat stress causes a substantial deleterious effect on dairy industry, including reducing growth, reproductive efficiency, and milk yield and quality. Thus, strategies to mitigate the effects of HS need to be developed. However, there is no single strategy that is used to mitigate the negative impacts of HS in dairy cattle farming. The following strategies can reduce the effects of HS in dairy cattle: (1) physical modification of the environment (shading, cooling), (2) genetic development of heat tolerant animals, and (3) improved feeding management strategies (65, 123–126). A combination of the above practices can enhance the production performances of lactating dairy cattle in hot and humid climates, especially in tropical and sub-tropical countries. It has been indicated that cooling technologies have a high positive impact on eliminating the adverse effects of high THI in dairy cows (1). However, those strategies do not completely alleviate the adverse effects of HS owing to animal-related factors such as genetics, hair coat and metabolic heat generated (127). As such, identification of dairy cattle resistant to high heat load and utilization of suitable mitigation strategies can lead to improvements in animal welfare and profitability (128).

This review provides a comprehensive description of the genes and genetic parameters for heat tolerance in dairy cattle breeds reared in tropical countries. It highlights various indicators used to measure HS response in dairy cattle including environmental variables such as ambient temperature, relative humidity, solar radiation and wind speed. It also discussed decline in milk yield and alterations in milk composition traits such as fats, proteins, lactose, solids-not-fat percentages, and somatic cell score as indicators of HS. Additionally, physiological parameters such as core-body temperature, respiration rate, rectal temperature, panting score, drooling score, heart rate, and pulse rate as well as behavioral changes are potential indicators of HS in dairy cattle. This review describes various indicine and taurine dairy breeds reared in tropical countries and the candidate genes for heat tolerance that have been identified in tropical dairy cattle breeds. It discusses three approaches, namely gas and liquid-chromatography tandem mass spectrometry, GC-LC/LC–MS, and proton nuclear magnetic resonance (¹H-NMR), used to identify milk and blood metabolites as vital biomarkers of HS in dairy cattle breeds, and the effects of HS on those metabolites.

The phenotypic response and genetic parameters for heat tolerance in dairy cattle breeds reared in Asia, Latin America, and sub-Saharan countries are also presented with various genetic models used for estimating genetic components for heat tolerance. The review emphasized the genetic antagonism between milk production traits and heat tolerance, indicating the need to consider the genetic sensitivity of potential dairy cattle breeds reared in tropical environments to avoid the adverse effects of HS. The review also highlighted the potential use of Bos indicus breeds, which harbor a large number of genomic regions and candidate genes responsible for climatic adaptation, for crossbreeding with Bos taurus breeds to improve milk production and genetic adaptation in tropical countries. Management practices for mitigating HS, such as environmental modifications and improved nutrition and feeding strategies, were discussed, with genetic selection for heat tolerance being recommended as a cost-effective approach.

In conclusion, the adoption of genetic selection for heat-tolerant animals, along with environmental modifications, improved nutrition, and crossbreeding of temperate dairy breeds with tropical breeds, is recommended for mitigating the impacts of HS in dairy cattle, particularly in tropical countries. Further research on candidate genes and genetic parameters associated with heat tolerance, as well as the development of appropriate genetic evaluation models, can contribute to the breeding of heat-tolerant dairy cattle breeds, ultimately improving their performance and resilience in the face of climate change. Crossbreeding of temperate breeds with local breeds that are resilient to HS is a quick way for improving thermotolerance of animals and thus, it is highly recommended in sub-Saharan Africa.

VH conducted the research during his PhD studies and prepared the initial draft of the manuscript. AN, ZN, CE-D, GM, RM, and SC conceived the study idea, agreed on the content, and edited the manuscript. All authors contributed to the article and approved the submitted version.

This study was funded by the Partnership for Skills in Applied Sciences, Engineering and Technology (PASET) through the Regional Scholarship and Innovation Fund (RSIF) awarded to VH to carry out doctoral studies at SACIDS Africa Centre of Excellence for Infectious Diseases, SACIDS Foundation for One Health, Sokoine University of Agriculture, Morogoro, Tanzania and International Livestock Research Institute (ILRI, Nairobi, Kenya). The funders had no role in study design, the decision to publish, or the preparation of the manuscript. The findings and conclusions of this study are those of the authors and do not necessarily represent the views of the funders.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.