Studies on Diagnosis of Bovine Brucellosis by Immunocyto-chemistry and Immunohistochemistry

Brucellosis is a zoonotic disease and an important contagious bacterial infection of livestock worldwide. The disease leads to innumerable loses in health and productivity. The disease is still endemic in various regions of the world including Latin America, Middle East, Africa, Asia and the Mediterranean basin (Pappas et al., 2006).
       
The losses are in the terms of reproductive efficiency, restriction on livestock commerce and animal replacement costs. On the other hand, Brucellosis is also a widespread zoonotic threat transmitted to humans by animals through direct contact with infected materials or indirectly by ingestion of animal products and by inhalation of airborne agents. Complete eradication has not been achieved till now.
       
Brucellosis is caused by facultative intracellular Gram-negative bacteria, nonspore-forming and non-capsulated bacteria which fall under the genus Brucella, a member of the α-2 Proteobacteria (Garrity, 2001). B. melitensis, B. abortus and B. suis are the most virulent species for domestic animals and humans. Brucellosis affects cattle, buffalo, sheep, goats, pigs and the disease is characterized by reproductive failure, particularly abortion in females (in the 3^rd trimester of pregnancy) and orchitis with sterility in males. Other manifestations of the disease include stillbirths, reduced milk yield, high frequency of retained placenta, prolonged calving interval and hygroma in knee joints (Radostits et al., 2000).
       
The isolation of Brucella from host tissues, milk, vaginal exudates etc. continues to be the “gold standard”, followed by bacteriological characterization. However, it is time consuming and has low sensitivity because Brucella are fastidious micro-organisms that can easily be overgrown by other contaminating bacteria. Moreover, the technique is less sensitive in chronic infections. More significantly, Brucella organisms are class III pathogens which require to be handled with utmost care and the isolation must be performed by person with sound technical skill.
       
To overcome these adversities, visualization of antigen antibody interaction assays have been tried for safe and timely diagnosis ahead of conventional isolation. Immunocytochemistry and Immunohistochemistry techniques have been used to detect Brucella abortus antigen and it is mainly based upon staining with species specific monoclonal antibody. Therefore, the main objective of the current study was to test the efficacy of Immunocytochemistry and Immunohistochemistry on Brucella isolates and on the clinical samples directly to get rapid results in diagnosis.

A total of 50 samples consisting of uterine discharges, vaginal mucus, fetal stomach contents and placenta were collected from cattle and buffaloes suffering from abortions and other reproductive disorders in and around Ludhiana, Punjab. The study was carried out during the period of February 2017 to July 2018, in the Department of Veterinary Microbiology and Department of Veterinary Anatomy, Guru Angad Dev Veterinary and Animal Sciences University, Ludhiana.
       
The samples were inoculated on BSM (Brucella specific medium) and the plates were incubated at 37°C under 5-10% CO₂ for up to 3-5 days and observed the growth. The identification of isolates was done based on morphology, cultural characteristics, different biochemical tests like oxidase, catalase, urease, nitrate reduction, H₂S production and indole, growth in the presence of dyes i.e. thionin and basic fuchsin.
       
DNA was isolated using hot cold lysis method. The Brucella isolates were confirmed by genus specific PCR primers B4/B5 (Baily et al.,1992) (Table 1). The contents and conditions of PCR are given in Table 2 and 3 respectively.
 

 

 

       
B. abortus reference strain S99 and Brucella isolates obtained from the clinical samples and the clinical samples directly were subjected to Immunocytochemistry in the present study.
 
Immunocytochemical studies
 
The polymer based horse radish peroxidase method used for immunocytochemical studies as per the procedure described earlier by Pathak et al., (2019). Bacterial cultures were suspended in PBS (pH 7.4). Thin smears were prepared on super frost positively charged slides (Biogenex). Smears were fixed by placing them in acetone at -20°C for 20 min. After fixation, heat induced antigen retrieval was done in citrate buffer-based antigen unmasking solution (H-3300, Vector Laboratories, USA) and heating in microwave at 98°C for 10 minutes. Slides were then left for 30 min in hot antigen retrieval solution and washed in 0.1M phosphate buffered saline (at pH 7.4). The endogenous peroxidase activity was blocked by immersing the sections in 3% (v/v) H₂O₂ in methanol for 20 min followed by washing in 0.1M phosphate buffered saline (at pH 7.4). To prevent nonspecific binding of antibodies sections were blocked with normal horse serum (Vector’s Laboratories, USA). The sections were incubated with primary antibody [Brucella abortus, BA 35 (Santa Cruz)], at 4°C for overnight in the humid incubation chamber. After washing in 0.1M phosphate buffered saline (at pH 7.4), the sections were incubated with universal secondary antibody (Vector Laboratories, USA). The chromogen used was 3, 3’-diaminobenzidine tetrahydro- chloride (DAB) (Vector Laboratories, USA) with Gill’s III hematoxylin counterstaining. The sections were washed in running tap water, dehydrated, cleared and mounted with DPX.
 
Immunocytochemistry on clinical samples
 
Fetal stomach contents and uterine discharges were centrifuged at 5000 rpm for 5 minutes. The pellet was then resuspended in PBS (pH 7.4). The contents were centrifuged again at the same speed to acquire the pellet by discarding the supernatant. This process was repeated four times.
       
Immunocytochemistry was carried out using the same protocol mentioned in the Section above.
 
Immunohistochemical studies
 
Protocol for tissue processing for paraffin sectioning
 
Placental cotyledons were collected from the aborted fetuses from different regions (six regions randomly) of placentomes and were processed for paraffin sectioning for immunohistochemical studies as given below:
 
➢  10% neutral buffered formalin for 12-24 hours
➢  70% Alcohol                      1 hour
➢  80% Alcohol                      1 hour
➢  90% Alcohol                      1 hour
➢  100% Alcohol I                  1 hour
➢  100% Alcohol II                 1 hour
➢  Acetone I                           15 min
➢  Acetone II                          15 min
➢  Benzene I                          15 min
➢  Benzene II                         15 min
➢  Paraffin Wax I                    2 hr
➢  Paraffin Wax II                   Overnight
➢  Paraffin Wax III                  2 hr
➢  Block preparation with paraffin.
➢  Sectioning (4-5 µm thickness) with microtome (Leica).
➢  Picking of sections on grease free glass slides.
➢  Drying of sections by incubating them in incubator at 60°C for one hour.
 
Immunohistochemical studies on placental tissue
 
Five-micron sections were cut and mounted on super frost positively charged slides (Biogenex). Sections were deparaffinized (Xylene) and rehydrated to water (descending grades of alcohol). After fixation, heat induced antigen retrieval was done in citrate buffer-based antigen unmasking solution (H-3300, Vector Laboratories USA) and heating in microwave at 98°C for 10 minutes. Slides were then left for 30 min in hot antigen retrieval solution and washed in 0.1M phosphate buffered saline (at pH 7.4). The endogenous peroxidase activity was blocked by immersing the sections in 3% (v/v) H₂O₂ in methanol for 20 min followed by washing in 0.1 M phosphate buffered saline (at pH 7.4). To prevent nonspecific binding of antibodies sections were blocked with normal horse serum (Vector’s Laboratories, USA). The sections were incubated with primary antibodies (Brucella abortus, BA 35, Santa Cruz) at 4°C for overnight in the humid incubation chamber. After washing in 0.1M phosphate buffered saline (at pH 7.4), the sections were incubated with universal secondary antibody (Vector Laboratories, USA). The chromogen used was 3, 3’-diaminobenzidine tetrahydrochloride (DAB) (Vector Laboratories, USA) with Gill’s III haematoxylin counterstaining. The sections were washed in running tap water, dehydrated, cleared and mounted with DPX.
 
Image analysis
 
The immunostained stained slides were imaged using Nikon microscope (80i) with photographic unit. The images were evaluated for presence of Brucella which was visualized by presence of brown coloured DAB reaction.

Isolation
 
A total of four isolates of Brucella were recovered from a total of 50 samples. These isolates were confirmed based on the morphological and biochemical characteristics and by genus specific PCR. The samples collected across the cases were not homologous due to the unpredictable nature and timing of the abortion. Fetal stomach contents could be aspirated from some cases, whereas in other cases they could not be aspirated in present study. In such cases placenta and other reproductive discharges were collected.
       
After an incubation of 3-5 days, pinpoint, smooth, glistening, bluish, translucent colonies were noticed which upon ageing became opaque. Gram’s staining revealed gram negative, coccobacillary organism or rods whereas by modified Ziehl-Neelsen staining they appeared red with blue background. The organism failed to grow on McConkey’s lactose agar (MLA) and lacked motility. The isolates were subjected to various biochemical tests like H₂S, urease, nitrate reduction and indole. All the four isolates were found positive for catalase, oxidase, urease, H₂S production and nitrate reduction test and negative for indole test. DNA from B. abortus S19 reference strain as well as from all the four Brucella isolates was subjected to PCR using Brucella genus specific primer pair B4/B5 targeting bcsp31 gene (Baily et al., 1992). An amplicon size of 223 bp was detected in all the four isolates as well as in B. abortus S19 reference strain.
 
Detection by immunocytochemistry and immuno- histochemistry
 
All the 4 clinical isolates subjected to immunocytochemistry yielded positive results with DAB chromogen exhibiting small, coccobacillary organisms (Fig 1). Very few reports are available in literature relating to use of Immunocyto- chemistry technique to identify Brucella spp from clinical samples. However, Mohammadi and Golchin (2017) developed an Immunofluorescence test based on similar technique against OMP by tagging OMP 16 with FITC.
 

       
The present study depicts the efficiency of immunocyto-chemical detection of Brucella directly in the clinical sample. A total of 43 samples comprising of fetal stomach contents, uterine discharges were collected for isolation were processed and subjected to immunocytochemistry of which a total of five clinical samples comprising of fetal stomach contents (11.62%) were positive (Fig 2). Of the five clinical samples that were positive by immunocytochemistry, only three isolates (60%) could be recovered on culture suggesting the greater sensitivity of immunocytochemistry when compared to isolation. No DAB reaction was observed in the negative control which consists of primary and secondary antibody incubation without Brucella culture (Fig 3). Larsen et al., (2015) verified intracellular presence of Brucella pinnipedialis using Immunocytochemistry to substantiate the finding that seals may not be the primary host of B. pinnipedialis. There are no prominent immunocytochemical studies on diagnosis of Brucella until now but other studies pertaining to various bacteria regarding direct visualization of organism or their products in clinical specimens by staining with species specific monoclonal antibody. They concluded that immunocytochemistry is sensitive, specific and a better technique than Nucleic acid amplification and can be used as an adjunct to conventional morphology (Goel et al., 2012).
 

 

       
Histopathological examinations done using paraffin sections stained with haematoxylin and eosin stain revealed that placental cotyledons showed chorionic villi lined with trophoblastic cells, binucleated cells and necrotic areas (Fig 4). Immunohistochemical (IHC) staining of placental sections demonstrated numerous labeled bacteria in trophoblasts (Fig 5 and 6). Thus, immunohistochemical examination of paraffin wax-embedded tissues for B. abortus antigens was found to be both sensitive and specific. It is, therefore, capable of demonstrating the distribution of organisms in the tissues, a valuable for the study of pathogenesis of B. abortus infection (Meador et al., 1986, 1988; Hong et al., 1991; Perez et al., 1998) and also for diagnosis. The intracellular localization of Brucella in the placental trophoblast has been reported in the placenta of bovine (Meador and Deyoe, 1989). Trophoblasts are thought to be the primary target cell for invasion and multiplication of B. abortus in the placenta (Anderson et al., 1986).  The tropism might be due to the presence of erythritol, or due to hormone synthesis by trophoblastic cells (Samartino and Enright, 1993).
 

 

       
In the present study out of the seven, three (42.85%) placental cotyledons were positive by immunohisto- chemistry. However only one (14.28%) isolate could be recovered on culture from 7 samples of placental cotyledons suggesting that immunocytochemistry is sensitive test in comparison to conventional isolation of bacteria. A lowered recovery rate on culture might be due to degenerated microorganisms, improper handling during isolation, lack of expertise and delay in processing of samples leading to death of fastidious Brucella organism. Positive Immunoreactivity was detected in higher number of samples which however were not always positive. It was concluded that immunohistochemistry technique was a sensitive technique for detecting B. abortus antigens in formalin-fixed tissues from naturally aborted bovine fetuses (Perez et al., 1998, Sözmen et al., 2004).

Isolation by way of culture continues to be a gold standard, but findings of present study reveal that there were certain instances where isolation failed to bring out the organism from the clinical sample rather the diagnostic tests like Immunocytochemistry and immunohistochemistry exhibited positive reaction suggesting better sensitivity than isolation. Diagnosis by immunocytochemistry is considered more safe and rapid as it reduces the exposure of the technician to the pure Brucella culture in isolation and saves time which is required in abundance for isolation and also that for suspected cases in which the bacteriologic culture is negative or the material is fixed in formalin, immunohisto- chemistry may be used as a diagnostic tool for the detection of Brucella organism.