Anaplasma ovis as the suspected cause of mortality in a neonatal elk calf

Abstract

Anaplasma ovis infection is known to occur in elk experimentally, but without clinical signs or significant clinicopathologic changes. An elk farm in southern Indiana experienced the death of 3 neonates. Gross findings suggested hemolytic anemia as the cause of death. Splenic impression smears revealed numerous intra-erythrocytic parasites compatible with Anaplasma spp. Products of a semi-nested PCR targeting the msp4 gene of A. ovis were sequenced and had 100% identity with published A. ovis sequences. Given the clinical presentation, vertical transmission of A. ovis was suspected. Pathologic and molecular findings confirmed that natural A. ovis infection occurred in an elk calf.

In August 2015, an elk farm in southern Indiana experienced a bout of neonatal mortality. Three ~5-d-old calves died within a short time frame, each with dyspnea and weakness. Eleven adult elk, including 2 bulls and 9 cows, were present on the farm. The submitting farmer suspected poor milk production from the cows and had been supplementing with sheep or goat milk. An autopsy was conducted on a 5-d-old bull calf that weighed ~20 kg. The calf was thin and had pale mucous membranes. Blood that oozed from cut surfaces during the postmortem examination was watery (Fig. 1A), and urine was dark red (Fig. 1B). The spleen was enlarged and firm on cut section. The rumen was almost empty; the abomasum contained gray liquid.

Figure 1.

Anaplasma ovis infection in a neonatal elk calf. A. Watery blood from the affected elk calf observed during the postmortem examination. B. Urine, free in the abdomen from the opened urinary bladder, was dark red (asterisk). C. A splenic impression smear revealed intraerythrocytic organisms in 70–80% of erythrocytes. Wright–Giemsa. D. Parasitized erythrocytes were present within macrophage cytoplasm (arrow). Wright–Giemsa.

In a splenic impression smear stained with a commercial Wright–Giemsa stain (VWR, Radnor, PA), 70–80% of examined erythrocytes contained 1–4, intraerythrocytic, dark blue-to-purple, round inclusions (Fig. 1C, D). Inclusions were located away from the margins of the cell and, in some cases, within the center of the erythrocyte. Phagocytosis of parasitized erythrocytes was observed in the impression smear (Fig. 1D); erythrophagocytosis was common in a histologic section of the spleen. Hepatic lipidosis was also observed histologically. No other significant histologic lesions were evident.

Anaerobic culture of the liver and PCR testing for Leptospira sp. were negative. Toxic causes of hemolysis were considered unlikely, given that the calf and dam were not treated with drugs or compounds known to be hemolytic. Ingestion of a sufficient amount of plants with hemolytic properties was considered unlikely, given the age of the calf and a rumen devoid of a significant amount of plant material.

Given the clinical signs, gross lesions of anemia, and the morphology of intraerythrocytic organisms observed on the splenic impression smear, parasites of red blood cells such as Anaplasma sp. were suspected. Infection with A. marginale in elk has been documented previously, although the infection was considered subclinical.¹⁷ Clinical anaplasmosis in cattle as a result of infection with A. marginale is a common finding in southern Indiana; however, a duplex PCR assay performed at the Kansas State Veterinary Diagnostic Laboratory for A. marginale and A. phagocytophilum on sections of spleen from the elk calf was negative. Sanger sequencing of 16S ribosomal RNA gene amplicons from the spleen of the infected elk calf was uninformative because of mixed environmental sequences. A. ovis was investigated next. Because the majority of published A. ovis sequence data are based on the msp4 gene, a semi-nested PCR targeting msp4 was performed, and the product was then cloned into pCR4-TOPO vector (Thermo Fisher Scientific, Waltham, MA). Primary PCR utilized previously developed primers MSP45F (5’-GGGAGCTCCTATGAATTACAGAGAATTGTTTAC) and MSP43 (5’-CCGGATCCTTAGCTGAACAGGAATCTTGC); and the semi-nested reaction used MSP45 and AovisMSP4Rev (5’-GAGTAATTGCAGCCAGGGACTCT).⁴ The latter primer was from a study that was specifically designed to differentiate between A. ovis and A. marginale.¹¹ Amplification reactions contained 12.5 µL of GoTaq (Promega, Madison, WI), 1 µL of 25 mM MgCl₂, 2 µL of primers (10 µM F primer + 10 µM R primer), and 3 µL of DNA at ~30 ng/µL. Cycling conditions included an annealing temperature of 60ºC and an extension time of 1 min. The nested product was 470 bp in length. Six clones were sequenced, and 2 different sequence variants were obtained (5 of type 1 [GenBank accessions MH908095–MH908099] and 1 of type 2 [GenBank accession MH908094]; Fig. 2). The Msp4 type 1 sequences had 100% identity with A. ovis Msp4 sequences in GenBank from a variety of sources, including mule deer (100%, GenBank accessions ABD48216 and DQ674249) from California and Montana.¹³ Msp4 from A. marginale and A. centrale differed from the sequences derived from the infected elk, with 90–97% and 88% identity, respectively (Fig. 2).

Figure 2.

Msp4 sequence alignments. Deduced amino acid sequences obtained after sequencing semi-nested PCR products targeting the msp4 gene are shown along with the same region from full-length sequences from Anaplasma ovis (Ao), A. centrale (Ac), and A. marginale (Am) Msp4. The numbers above the sequence refer to the amino acid number in the full-length version of A. ovis Msp4. The full-length sequences were obtained from GenBank, and their accessions are given. Two variants of the msp4 gene were obtained; both had 100% identity to published A. ovis Msp4 sequences.

A. ovis is a well-known hemoparasite of sheep and goats throughout the world, with infections documented in Africa, Asia, Mediterranean Europe, and North America.^6,9 Infection is usually considered either mild or subclinical in domestic small ruminants, with the most severe clinical signs of disease manifesting as emaciation and anemia in the Mediterranean region.¹² As with many diseases, concurrent illnesses or inadequate husbandry in sheep and goats may cause overt manifestations of anaplasmosis.⁹ In certain wildlife species, including bighorn sheep within the western United States, and Mongolian reindeer, severe disease secondary to infection with A. ovis has been described.^5,10 In addition, several wild ruminant species found in North America, including pronghorn antelope, white-tailed deer, mule deer, and elk, have all been infected experimentally with A. ovis, demonstrating the wide host range of this hemoparasite.^3,14,15,17 Although experimental infection of elk with A. ovis has been confirmed previously,¹⁷ the elk in that study showed no clinical signs and no decrease in hematocrit. In contrast to the cited study, in which the elk were 8-mo-old, the naturally infected animals in our study were neonates.

Given the age of the infected elk in our study and the severity and course of disease, infection of the fetus in utero was considered likely. Vertical transmission of the related A. marginale in cattle is well documented.^1,2 In addition, transplacental transmission of A. ovis was demonstrated experimentally in 3 of 16 sheep fetuses or neonates born to ewes infected with the bacteria at various times during gestation.¹⁶ Thus, natural in utero transmission of this hemoparasite in elk is considered likely. Young age or fetal status at the time of infection may be the predisposing factor that allowed the development of disease with this otherwise subclinical infection in elk.

Many of the observed gross anatomic, cytologic, and histologic lesions were consistent with anaplasmosis in our case; however, hemoglobinuria indicated intravascular hemolysis. Anemia in clinical cases of anaplasmosis occurs secondary to extravascular hemolysis; hemoglobinuria is not considered common in Anaplasma spp. infection. Thus, the presence of hemoglobinuria was unusual. In our case, the farmer bottle-fed sheep or goat milk to supplement what was thought to be poor milk production by the elk cows. If colostrum from ewes or does infected with A. ovis was fed to the infected elk calves, perhaps a condition similar to neonatal isoerythrolysis, with profound intravascular hemolysis, was initiated. In case reports of clinical neonatal anaplasmosis in cattle as well as in descriptions of experimentally induced neonatal anaplasmosis in cattle, hemoglobinuria or hemoglobinuric nephrosis were not mentioned.^2,7

Another notable difference between our case and cases of neonatal anaplasmosis in other species was the high percentage of infected erythrocytes. In cases of natural and experimentally induced neonatal anaplasmosis in cattle, including experiments with A. centrale, parasitized erythrocytes ranged from <1% to 43%.^2,8 In the single reported experimental induction of ovine neonatal anaplasmosis using A. ovis, parasitemia was not detected in any of the neonatal lambs or fetuses born to ewes infected during gestation, and vertical transmission was only confirmed via transfer of neonatal or fetal blood to splenectomized lambs.¹⁶ Interestingly, in our case, 70–80% of erythrocytes observed on a splenic impression smear were infected with intra-erythrocytic parasites. Some authors have suggested that the timing of acute infection is important in the development of fetal infection and, ultimately, neonatal illness.⁸ For example, if acute infection occurs in the last two-thirds of pregnancy, clinical manifestation of disease in the neonate is more likely.⁸ Thus, the timing of acute infection in the elk cows may have been in the optimal window to allow for severe fetal infection and, ultimately, neonatal disease. The adult elk were never reported to be ill and, unfortunately, samples were not obtained from the elk cows for testing.

Our report documents that natural A. ovis infection can occur in elk, and this agent can cause severe clinical disease in neonatal elk calves. Notable clinicopathologic features of this case included the high number of infected erythrocytes observed on the splenic impression smear, as well as gross evidence of hemoglobinuria. Whether infection with A. ovis represents an important cause of neonatal mortality in free-ranging elk is unknown. Elk should be considered a possible wildlife reservoir for A. ovis. Previous work demonstrated susceptibility of elk to A. marginale; whether neonatal anaplasmosis is caused by this agent is unknown, but seems likely, given our case.¹⁷ Anaplasmosis should be considered as a cause of hemolytic anemia in elk calves.