**5.2.1 Clinical findings**

Host responses in the three experimentally infected sheep groups were first compared mainly according to typical high fever periods, microscopic observation and haematological values. *A. ovis* began to appear in the blood a week before the fever and the following records showed that the maximum of erythrocytes parasitized by *A. ovis* in any case did not exceed 2%.

All the animals developed the disease (Table 8) but symptoms varied in terms of severity and duration and none died. Fever syndrome (listlessness, anorexia, weakness, ruminal stasis, respiratory distress, increased heart and respiratory rates) and pallor of the mucous membranes were recorded in seven of the Suffolk group, in only one of the Comisana group and in none of the Altamurana.

*A. ovis* was isolated from one of the above splenectomized sheep. Lina was the donor for the 2009 lambs and Gilda for the 2010 lambs. Parasite density was estimated on thin blood film obtained by the buffy coat method and expressed as the percentage of parasitized red blood cells. At the peak of parasitaemia in the donor sheep (36% and 60% of red blood cells parasitized respectively), about 400 ml of blood were obtained and each lamb in the breed

Clinical evaluation was done on a daily basis and rectal temperatures were recorded every morning for 8 weeks post infection. Blood and serum samples were collected twice a week during the observation period. Haematological variables were evaluated using a haematology

The reversible switch from haemoglobin A to C was observed in the above HBBAB Suffolk lambs. The expression of the silent gene encoding for Hb C was detected by isoelectric focusing and quantified by high performance liquid chromatography (Alloggio *et al.*, 2009).

First, differences between breed groups for clinical and haematological data were assessed using analysis of variance (ANOVA) by GLM procedure (SAS, 1990). A second ANOVA was carried out only for the Altamurana group, considering the interaction between the alpha globin type (2 levels: NH and EH) and each clinical variable. The last ANOVA was carried out only for the Suffolk group considering the interaction with the beta globin type (2 levels: AB and BB) of the linear and quadratic regression of each haematological variable on the number of days from the infection, with three of the Suffolk that were AB compared to as many BB. This analysis was performed for the Suffolk, where, as cited above, both AB

and BB genotypes were found, whereas only BB animals occurred in the Comisana.

The following is an overall picture of the findings where they are reported and discussed

Host responses in the three experimentally infected sheep groups were first compared mainly according to typical high fever periods, microscopic observation and haematological values. *A. ovis* began to appear in the blood a week before the fever and the following records showed that the maximum of erythrocytes parasitized by *A. ovis* in any case did not

All the animals developed the disease (Table 8) but symptoms varied in terms of severity and duration and none died. Fever syndrome (listlessness, anorexia, weakness, ruminal stasis, respiratory distress, increased heart and respiratory rates) and pallor of the mucous membranes were recorded in seven of the Suffolk group, in only one of the Comisana group

groups was inoculated intraperitoneally with 25 ml of infected blood.

analyzer. Parasite density was estimated on thin blood film as above described.

**5.1.3 Experimental infection** 

**5.1.4 Clinical observations** 

**5.1.5 Haemoglobin phenotype** 

**5.2 Results and discussion** 

**5.2.1 Clinical findings** 

exceed 2%.

relating to the different approaches.

and in none of the Altamurana.

**5.1.6 Statistics** 


Table 8. Overview of responses to anaplasmosis in the three analyzed breeds.

The haematological patterns were then analyzed in detail comparing the intra breed variations between the different physiopathological moments – normal health status (time 0=T0), acute phase (time 1=T1 ) recovery phase (time 2=T2 ) - and the between breed variations intra physiopathological moments (table 9). Finally, clinical parameters, such as incubation time (I.T) after infection, temperature peak (T.P.), percentage decrease in haematocrit ( HCT), percentage decrease in haemoglobin content ( Hb) expressed as gr Hb/dl blood, percentage decrease in red blood cells ( RBC) were evaluated for each breed (Table 10).


Table 9. Haematological parameters assessed for the three analyzed breeds, namely normal health status before infection (time 0=T0), during the acute phase (time 1=T1 ) and during the recovery phase (time 2=T2 ). Means within columns with different letters significantly differ: capital letters: P <0.01; small letters: P<0.05.

Tolerance to Tick-Borne Diseases in Sheep:

**Breed I.T** 

percentage decrease in red blood cells (RBC).

found.

(days)

Highlights of a Twenty-Year Experience in a Mediterranean Environment 469

Suffolk 24.2A 40.5 A 65.4A 60.2 A 60.3 A Comisana 38.8B 39.9 B 57.7 B 56.9A 55.9A Altamurana 25.3A 39.9 B 19.2 C 19.8 B 33.7 B

**HCT**  (%)

**Hb**  (%)

**RBC**  (%)

**T.P.**  (°C)

Table 10. Clinical parameters assessed for the three breeds analyzed at the peak of the disease, namely incubation time (I.T) after infection, temperature peak (T.P.), percentage decrease in haematocrit (HCT), percentage decrease in haemoglobin content (Hb),

As to the functional effect of beta globin genes on the recovery from anemia, Figure 6 shows the results of the analysis performed in the Suffolk, where both AB and BB genotypes were

AB genotype

BB genotype

Fig. 6. Trend of Haemoglobin (Hb, g/dl) as a function of the number of days from the infection (DAYS) for the haemoglobin genotypes AB and BB (Modified from Alloggio et al. 2009).

**5.2.2 Functional effect of beta globin genes on the recovery from anemia** 

As already reported, none of the animals had more than 2% erythrocytes parasitized by *A. ovis*. This confirms the role the spleen plays in the phagocytosis and clearance of parasitized erythrocytes; otherwise only splenectomized sheep showed significant percentages of parasitized erythrocytes (table 5).

Mean parasitaemia in the single group could theoretically be inferred from the RBC and conclude that the higher the RBC value, the higher the suceptibility of the erythrocytes to *Anaplasma* infection. According to the results shown in table 8 depicting the response of the three breeds to infection, broken down according to the dose of infection, symptoms, need for therapeutic intervention and expected mortality, seven out of the Suffolk group recovered after being treated every two days for a week with oxytetracycline and dexamethasone whereas seven subjects of the Comisana group recovered from clinical anaplasmosis with no drug treatment other than a single dose of dexamethasone. The highest degree of tolerance was observed in the Altamurana group where all the subjects showed only mild alteration of behaviour and basic life functions. Comparison of the haematological patterns of the three breeds at T0 in table 10 revealed that, in normal health conditions, the differences which may be noticed are consistent with those of earlier studies described in section 2, indicating that both environmental and productive specialization seem to account for the different physiological results.

Hence the Altamurana breed is characterized by significantly lower RBC and Hb values and by significantly higher MCV, MCH and MCHC values than Suffolk, a northern meat breed, while Comisana, a Mediteranean dairy breed has intermediate values. At T1 and T2 the same variation pattern may be observed, that is:


The results shown in table 10 confirmed the differences among the three breeds both in terms of quantitative (Altamurana *vs* Suffolk) and temporal (Comisana *vs* Suffolk) variation in haematological parameters. While Suffolk animals displayed the most severe reduction in the number of erythrocytes and haemoglobin content, Altamurana was characterized by a more controlled response, with only a minor and more gradual decline in RBC and Hb values. Comisana experienced a more severe reduction in the number of erythrocytes than did Suffolk, though the decline in RBC values was not accompanied by a decrease in the total haemoglobin content as severe as that observed in Suffolk. This could be due to the significantly higher MCH values characterizing the Comisana haematological pattern as compared to the Suffolk.

Considering the overall responses shown in table 8 and detailed in table 9 and 10, there is no doubt that we are dealing with very different animal groups exhibiting different physiopathological phenotypes where a healthy haematological picture plays a relevant role.

As already reported, none of the animals had more than 2% erythrocytes parasitized by *A. ovis*. This confirms the role the spleen plays in the phagocytosis and clearance of parasitized erythrocytes; otherwise only splenectomized sheep showed significant

Mean parasitaemia in the single group could theoretically be inferred from the RBC and conclude that the higher the RBC value, the higher the suceptibility of the erythrocytes to *Anaplasma* infection. According to the results shown in table 8 depicting the response of the three breeds to infection, broken down according to the dose of infection, symptoms, need for therapeutic intervention and expected mortality, seven out of the Suffolk group recovered after being treated every two days for a week with oxytetracycline and dexamethasone whereas seven subjects of the Comisana group recovered from clinical anaplasmosis with no drug treatment other than a single dose of dexamethasone. The highest degree of tolerance was observed in the Altamurana group where all the subjects showed only mild alteration of behaviour and basic life functions. Comparison of the haematological patterns of the three breeds at T0 in table 10 revealed that, in normal health conditions, the differences which may be noticed are consistent with those of earlier studies described in section 2, indicating that both environmental and productive specialization

Hence the Altamurana breed is characterized by significantly lower RBC and Hb values and by significantly higher MCV, MCH and MCHC values than Suffolk, a northern meat breed, while Comisana, a Mediteranean dairy breed has intermediate values. At T1 and T2 the same

 haematologically, *A. ovis* infection does not seem to seriously affect Altamurana whose response may be described as a moderate normochromic normocytic anemia followed by a normochromic macrocytic pattern representing an active regeneration

 conversely, the Suffolk and Comisana animals exhibited a violent response to *A. ovis* with a severe anaemia. The hyperchromic and macrocytic anaemia in the Suffolk was followed by a slow regeneration and the hypochromic normocytic anaemia of the Comisana by an active regeneration phase, similar to the Altamurana pattern, as

The results shown in table 10 confirmed the differences among the three breeds both in terms of quantitative (Altamurana *vs* Suffolk) and temporal (Comisana *vs* Suffolk) variation in haematological parameters. While Suffolk animals displayed the most severe reduction in the number of erythrocytes and haemoglobin content, Altamurana was characterized by a more controlled response, with only a minor and more gradual decline in RBC and Hb values. Comisana experienced a more severe reduction in the number of erythrocytes than did Suffolk, though the decline in RBC values was not accompanied by a decrease in the total haemoglobin content as severe as that observed in Suffolk. This could be due to the significantly higher MCH values characterizing the Comisana haematological pattern as

Considering the overall responses shown in table 8 and detailed in table 9 and 10, there is no doubt that we are dealing with very different animal groups exhibiting different physiopathological phenotypes where a healthy haematological picture plays a relevant

percentages of parasitized erythrocytes (table 5).

seem to account for the different physiological results.

variation pattern may be observed, that is:

documented by the high MCV values.

phase.

compared to the Suffolk.

role.


Table 10. Clinical parameters assessed for the three breeds analyzed at the peak of the disease, namely incubation time (I.T) after infection, temperature peak (T.P.), percentage decrease in haematocrit (HCT), percentage decrease in haemoglobin content (Hb), percentage decrease in red blood cells (RBC).

#### **5.2.2 Functional effect of beta globin genes on the recovery from anemia**

As to the functional effect of beta globin genes on the recovery from anemia, Figure 6 shows the results of the analysis performed in the Suffolk, where both AB and BB genotypes were found.

Fig. 6. Trend of Haemoglobin (Hb, g/dl) as a function of the number of days from the infection (DAYS) for the haemoglobin genotypes AB and BB (Modified from Alloggio et al. 2009).

Tolerance to Tick-Borne Diseases in Sheep:

<sup>N</sup> T.P. (°C)

> 40.1 A ± 0.18

39.8 B ± 0.12

PCV (g/dl)

> 25.9 ± 1.55

> 26.3 ± 1.18

factor in checking the results.

the results.

P <0.01.


NH 9

EH 9

than in the NH individuals.

**6. Conclusions** 

Highlights of a Twenty-Year Experience in a Mediterranean Environment 471

The second relevant element was the level of response which undoubtedly is a limiting

Thirdly, owing to experimental constraints, the sample size was of nine subjects per each alpha haplotype group. Hence, based on these considerations, we could not expect striking results as to the functional effect of the alpha globin gene arrangements, except in the case of a strong interaction with the response to experimental infections. Despite our hopes, our concerns were well-founded because all the above elements of prejudicial questions affected

> Hb (g/dl)

> > 8.1 ± 0.89

8.5 ± 0.39

Table 11. Temperature peak and haematological parameters assessed during the acute phase of the disease, for the two Altamurana groups classified on the basis of the -globin gene arrangement. Means within columns with different letters significantly differ; capital letters:

Altamurana subjects exhibited a very mild symptoms and no patent differences could be recorded in terms of haematological pattern between the EH and NH individuals within the Altamurana group. The EH group had a temperature peak that was significantly lower (P<0.001) than that of the NH group. This suggested that the level of response to infection in the EH group was lighter than in the NH group (Table 12). Morever, as shown in table 11, though no significance was attained by the ANOVA when the mean values of the haematological parameters of the two groups were compared, a univocal trend emerged whereby the RBC, PCV and Hb values in the EH group decreased less than in the NH group. These two phenomena seem to indicate a milder *Anaplasma* infection in EH subjects

Several examples of breed-related tolerance to diseases have been reported worldwide but often the claims made for specific breeds have not been subject to scientific investigation. As to small ruminants, only tolerance to Heartwater (Cowdriosis) has been documented for Damara, a South African native sheep breed (Commission on Genetic Resources for Food and Agriculture, 2007). This report extends our knowledge about tolerance to tick borne

diseases. The main findings can be summarized in the following points:

RBC (106/l)

> 7.8 ± 0.85

8.0 ± 0.31 MCV (fl)

> 33.1 ± 1.03

> 33.7 ± 0.73

MCH (pg)

10.4 ± 0.51

10.6 ± 0.43 MCHC (g/dl)

> 31.4 ± 0.51

> 31.4 ± 0.47

WBC (103/l)

> 10.3 ± 1.29

> 10.5 ± 1.17

Goats and some sheep under conditions of erythropoietic stress (anaemia) or hypoxia, synthesize a juvenile haemoglobin (Hb) type, Hb C, where -globin is encoded by the silent gene *HBBC*. Anaemia causes a change in the type of circulating haemoglobin only in sheep carrying A-globin haplotype, where Hb A is replaced by Hb C. Pioneered by the work of van Vliet and Huismam (1964), the Hb C in Caprini species has been thoroughly studied and particularly the mechanism of reversible switching has triggered focused research in the 70's (Nienhuis and Anderson, 1972; Nienhuis and Bunn, 1974). Little information is available on the effect of Hb A replacement by Hb C. Owing to the high oxygen affinity of Hb C (Huisman and Kitchens, 1968), the reversible switch from Hb A to Hb C may be considered a way to cope with the reduced amount of oxygen available at higher altitudes and thus suggest a positive effect on the fitness of mountain Caprini. Conversely, in the case of erythropoietic stress, it was suggested (Pieragostini et al., 1994; Pieragostini et al., 2006) that Hb C might negatively affect peripheral oxygen delivery and worsen the clinical picture of sheep breeds native of areas with endemic haemotropic pathogens. Hence, in the present case, the *conditio sine qua non* for checking the effect of beta globin genotype was the detection of Hb C switched on in AB individuals following a strong erythropoietic stress. As an example, figure 6 also shows the different trend after infection (days = 0) of the Hb content for the AB and BB ewes. Apart from individual differences observed in reaching the lowest Hb values, the recovery was always faster in BB sheep, as also indicated by the significantly higher quadratic regression coefficient of Hb for BB *vs* AB genotype (0.0045 *vs* 0.0027, P < 0.03). The different behaviour of one of the three heterozygous subject (Figure 6) may be justified by its less severe haematological picture due to a lower anoxaemic stress, also confirmed by the lack of the Hb A to Hb C switch in the same animal.

#### **5.2.3 Functional effect of alpha globin system on the response to the anaplasmosis**

As mentioned before, the extra numeral alpha globin haplotypes (EH) were suggested to be related to the host's response to TBDs. The unusual frequency of EH recorded in southern Italian sheep breeds and the peculiar haematologic pattern exhibited by the EH homozygous subjects may be taken as evidence of a selective advantage of the corresponding phenotypes in endemic TBD areas. Individuals carrying extra alpha-globin genes exhibit an overall blood picture mimicking a thalassaemia-like syndrome. In greater detail, when the erythrocytes of EH homozygotes were compared with those of NH individuals, the former had fewer erythrocytes that were bigger in size and had a higher Hb content and a greater erythrocyte osmotic fragility. These changes in EH homozygotes were assumed to produce an unfavorable environment for the parasites.

Thus, the trial checked this hypothesis as the different haematological patterns and the accelerated turnover of erythrocytes of EH individuals compared to the NH ones were expected to produce differences in the spread of the pathogen into the host blood.

Unfortunately a relevant element of prejudicial questions to obtain maximum results was the fact that only EH heterozygotes were present in the Altamurana group, since no homozygous lambs were found during the population survey. In normal health conditions, such as those recorded in the experience reported by Pieragostini et al. (2003), the EH heterozygotes showed an intermediate pattern, between that of EH homozygotes and that of NH homozygotes (Pieragostini, unpublished data).

Goats and some sheep under conditions of erythropoietic stress (anaemia) or hypoxia, synthesize a juvenile haemoglobin (Hb) type, Hb C, where -globin is encoded by the silent gene *HBBC*. Anaemia causes a change in the type of circulating haemoglobin only in sheep carrying A-globin haplotype, where Hb A is replaced by Hb C. Pioneered by the work of van Vliet and Huismam (1964), the Hb C in Caprini species has been thoroughly studied and particularly the mechanism of reversible switching has triggered focused research in the 70's (Nienhuis and Anderson, 1972; Nienhuis and Bunn, 1974). Little information is available on the effect of Hb A replacement by Hb C. Owing to the high oxygen affinity of Hb C (Huisman and Kitchens, 1968), the reversible switch from Hb A to Hb C may be considered a way to cope with the reduced amount of oxygen available at higher altitudes and thus suggest a positive effect on the fitness of mountain Caprini. Conversely, in the case of erythropoietic stress, it was suggested (Pieragostini et al., 1994; Pieragostini et al., 2006) that Hb C might negatively affect peripheral oxygen delivery and worsen the clinical picture of sheep breeds native of areas with endemic haemotropic pathogens. Hence, in the present case, the *conditio sine qua non* for checking the effect of beta globin genotype was the detection of Hb C switched on in AB individuals following a strong erythropoietic stress. As an example, figure 6 also shows the different trend after infection (days = 0) of the Hb content for the AB and BB ewes. Apart from individual differences observed in reaching the lowest Hb values, the recovery was always faster in BB sheep, as also indicated by the significantly higher quadratic regression coefficient of Hb for BB *vs* AB genotype (0.0045 *vs* 0.0027, P < 0.03). The different behaviour of one of the three heterozygous subject (Figure 6) may be justified by its less severe haematological picture due to a lower anoxaemic stress, also confirmed by the lack of the Hb A to Hb C

**5.2.3 Functional effect of alpha globin system on the response to the anaplasmosis**  As mentioned before, the extra numeral alpha globin haplotypes (EH) were suggested to be related to the host's response to TBDs. The unusual frequency of EH recorded in southern Italian sheep breeds and the peculiar haematologic pattern exhibited by the EH homozygous subjects may be taken as evidence of a selective advantage of the corresponding phenotypes in endemic TBD areas. Individuals carrying extra alpha-globin genes exhibit an overall blood picture mimicking a thalassaemia-like syndrome. In greater detail, when the erythrocytes of EH homozygotes were compared with those of NH individuals, the former had fewer erythrocytes that were bigger in size and had a higher Hb content and a greater erythrocyte osmotic fragility. These changes in EH homozygotes were

Thus, the trial checked this hypothesis as the different haematological patterns and the accelerated turnover of erythrocytes of EH individuals compared to the NH ones were

Unfortunately a relevant element of prejudicial questions to obtain maximum results was the fact that only EH heterozygotes were present in the Altamurana group, since no homozygous lambs were found during the population survey. In normal health conditions, such as those recorded in the experience reported by Pieragostini et al. (2003), the EH heterozygotes showed an intermediate pattern, between that of EH homozygotes and that

expected to produce differences in the spread of the pathogen into the host blood.

assumed to produce an unfavorable environment for the parasites.

of NH homozygotes (Pieragostini, unpublished data).

switch in the same animal.

The second relevant element was the level of response which undoubtedly is a limiting factor in checking the results.

Thirdly, owing to experimental constraints, the sample size was of nine subjects per each alpha haplotype group. Hence, based on these considerations, we could not expect striking results as to the functional effect of the alpha globin gene arrangements, except in the case of a strong interaction with the response to experimental infections. Despite our hopes, our concerns were well-founded because all the above elements of prejudicial questions affected the results.


Table 11. Temperature peak and haematological parameters assessed during the acute phase of the disease, for the two Altamurana groups classified on the basis of the -globin gene arrangement. Means within columns with different letters significantly differ; capital letters: P <0.01.

Altamurana subjects exhibited a very mild symptoms and no patent differences could be recorded in terms of haematological pattern between the EH and NH individuals within the Altamurana group. The EH group had a temperature peak that was significantly lower (P<0.001) than that of the NH group. This suggested that the level of response to infection in the EH group was lighter than in the NH group (Table 12). Morever, as shown in table 11, though no significance was attained by the ANOVA when the mean values of the haematological parameters of the two groups were compared, a univocal trend emerged whereby the RBC, PCV and Hb values in the EH group decreased less than in the NH group. These two phenomena seem to indicate a milder *Anaplasma* infection in EH subjects than in the NH individuals.
