Detection of intracellular bacteria in feline inflammatory cardiac,hepatic and renal diseases: a retrospective study using fluorescent in situ hybridisation

Abstract

Objectives

The aims of the present study were to determine whether bacteria are present in feline cardiac, hepatic and renal tissues where inflammation has been identified and to compare the location of any bacteria with areas of inflammation within those tissues. Fluorescent in situ hybridisation (FISH) facilitates visualisation of intracellular bacteria in tissues. There is little research looking at the role of intracellular bacteria in inflammatory disease within feline medicine.

Methods

Study group (SG) cases were selected from Ross University School of Veterinary Medicine’s pathology archive for 2012–2022. A total of 23 cases fulfilled the inclusion criteria. Three sequential sections were assessed with FISH (using eubacterial and non-eubacterial probes) and haematoxylin and eosin staining. Control group (CG) cases were selected from the same archive (n = 6) where death was trauma related; no other disease states were noted and the same three tissues were available for testing. Known bacteria-positive sections were included with each batch of slides processed to confirm successful hybridisation.

Results

Of the SG cases, 52.2% (95% confidence interval [CI] 30.6–73.2) demonstrated bacteria within some or all tissues tested, and 78.3% (95% CI 56.3–92.5) demonstrated the presence of inflammatory cells (ICs) in one or more tissues. Of the IC-positive SG cases, 61.1% (95% CI 35.7–82.7) demonstrated bacteria using FISH; the presence of bacteria in either the liver or kidney was frequently associated with the presence of ICs in 77.7% (95% CI 40.0–97.2) and 80% (95% CI 28.4–99.5) of cases, respectively. Among these, IC distribution did not match bacterial distribution. Of the CG cases, 83.3% (95% CI 35.9–99.6) were negative for ICs. Notably, in the IC-negative CG cases, two (40%) were positive for bacteria using FISH (95% CI 5.3–85.3). The Pearson χ² test demonstrated a χ² value of 0.71 (P = 0.40).

Conclusions and relevance

Despite this pilot study being limited by a small sample size, bacteria were successfully detected within formalin-fixed paraffin-embedded samples of feline heart, liver and kidney. We demonstrated that bacteria may not co-locate with all instances of inflammation, suggesting the need for greater vigilance for the presence of fastidious bacteria and/or low-grade infection.

Keywords

FISH intracellular bacteria inflammatory disease heart liver kidney fluorescent in situ hybridisation formalin-fixed paraffin embedded tissue FFPE

Introduction

Fluorescence in situ hybridisation (FISH) has diverse applications in many different fields of medicine and research. When used for the detection of intracellular organisms, its successful implementation is dependent upon protocol optimisation to both the organism and tissue of interest.^1
–4 The advantage FISH provides over other techniques relates to information on spatial distribution, viability and the number of organisms present within a given sample.⁵ This allows for a greater understanding of how and where organism populations reside and how multiple organisms coexist and interact within a host.⁶

The discovery of Escherichia coli as the cause of canine granulomatous colitis by Simpson et al⁷ increased awareness of FISH within small animal veterinary medicine. Since then, FISH has been used to screen for bacterial aetiologies in a range of diseases within the gastrointestinal, urinary and integumentary systems of small animals. There is currently little research looking at the role intracellular bacteria may have in disease states affecting other organ systems, especially within feline medicine.⁸

This retrospective observational pilot study aimed to use FISH to investigate the involvement of intracellular bacteria in inflammatory cardiac, hepatic and renal diseases within feline populations on Saint Kitts, West Indies.

The aims of the present study were to determine whether bacteria are present in feline cardiac, hepatic and renal tissues where inflammation/inflammatory diseases have been identified and to compare the location of any bacteria with histological areas of inflammation within those tissues.

Materials and methods

A search was conducted across electronic documents of case entries between 2012 and 2022 within the Ross University School of Veterinary Medicine (RUSVM) pathology archive, using the following terms: pericarditis, epicarditis, myocarditis, endocarditis, nephritis, cholangitis and hepatitis. Once completed, a further search was conducted using the term ‘itis’ to ensure all eligible cases were retrieved. Cases identified were then cross-referenced against their electronic patient records. Inclusion criteria were case histories or histopathology reports suggestive of inflammation or inflammatory disease within one, some or all of the heart, liver and kidney(s) and, where relevant, viable tissue samples had been collected and were available as formalin-fixed paraffin-embedded (FFPE) blocks.

Of 76 cases initially identified, 23 matched the inclusion criteria for the study group (SG). Of these, 17 were male (11 castrated, six entire) and six were female (three spayed, three entire), aged between one week and 16 years (median age 2 years).

To determine the size of the control group (CG), power calculations were performed using a STATA 16.1 two-proportion test, with an alpha of 0.05 and a power of 0.8. With the assumption that CG cases would have no bacteria or inflammatory cells (ICs) within the selected tissues, a group of six cats was required. CG cases were selected from the same pathology archive, spanning the same timeframe. The inclusion criteria were as follows: a complete history and medical records available; deaths were either witnessed or discovered and presented still in rigor; there were no other aetiologies or disease states noted within the pathology report (specifically, no inflammatory lesions reported in the heart, kidney and liver); and appropriate tissues had been collected as listed within histology requests and available as FFPE blocks. The CG included three males (all entire) and three females (one spayed, two entire), aged between five days and two years (median age 4.5 months). Of these, four had histories stating trauma as the cause of death (two were witnessed events and presented shortly thereafter; two were unwitnessed events and presented in rigor) and two were euthanased after a dog attack. A review of haematoxylin and eosin (H&E)-stained heart, liver and kidney samples of each case (both groups) was conducted using brightfield microscopy to ensure tissue viability.

For each case (in both the SG and CG), three consecutive sections were cut from each block; two to be used for FISH (for eubacterial and non-eubacterial probes, respectively) and one for H&E staining. These were mounted on 25 × 75 × 1.0 mm Probe-On-Plus microscope slides (Gray Fisher Healthcare; Thermo Fisher Scientific).

To mitigate the risk of cross-contamination during sample processing, sectioning and handling, all instruments and surfaces were thoroughly disinfected between cases using validated biocidal agents. Disposable tools were used whenever feasible, and reusable instruments were autoclaved or chemically sterilised as part of our standard operating procedures within our BSL-2 lab working environment.

Cy3 labelled 16S ribosomal RNA (rRNA) eubacterial probes (EUB338) with the sequence 5’-GCTGCCTCCCGTAGGAGT-3’, produced by Sigma-Aldrich (SA) (Merck) were used in this study. Non-eubacterial probes (NONEUB338), also labelled with Cy3 and from SA, were applied to a second sequential slide as a negative control wherever positive eubacterial results were detected. In addition, batch control slides (BCS) containing feline intestinal or liver tissue, in which bacteria had previously been identified, were included in each batch of slides tested to verify effective processing. The protocol used for this study is detailed in Figure 1.

Figure 1

Flow diagram detailing the fluorescent in situ hybridisation protocol developed for this study, which combines elements from protocols by the University of Texas at Dallas,⁹ Sigma-Aldrich¹⁰ and Cambio¹¹

All observations were performed by one individual (MR). Samples were deemed positive for bacteria where five or more organisms were identified within close proximity to one another, where signal output allowed the organisms to be easily identified, and where organisms were within the same visual plane as the surrounding tissues and were not located on the outermost surfaces of the sample.

The H&E-stained slides were then reviewed under brightfield microscopy by a board-certified pathologist (PB) to identify areas of inflammation. Inflammation was semi-quantitatively assessed based on histological pattern, cell type (neutrophilic, lymphoplasmacytic, plasmacytic or a combination) and distribution. The following criteria were applied to determine inclusion in the study group. For the heart, liver and kidney, inflammation was classified as focal or multifocal and at least minimal (defined as at least five inflammatory cells per high-power field at 40 × objective), with or without associated tissue damage. Some individual inflammatory cells were considered incidental, whereas clusters or diffuse infiltrates accompanied by tissue damage were included. These areas were marked on the underside of the slide with an indelible fine-point marker for comparison with regions identified as positive for bacteria.

Percentages were reported and the respective 95% confidence intervals (CIs) were calculated using the Clopper–Pearson method. The proportion of bacteria-positive cases (SG vs CG) was compared using a Pearson χ² test, with significance taken at P = 0.05. The analysis was performed using STATA 16.1.

Results

A summary of organ-specific bacterial detection rates compared with IC detection rates is presented in Table 1.

Table 1

Summary of organ-specific bacterial detection rates vs inflammatory cell (IC) detection rates

	Study group		Control group
	Bacteria positive	IC positive	Bacteria positive	IC positive
Heart	8	2	1	0
Liver	10	13	1	1
Kidney	5	12	1	0

Some cases were positive for bacteria and/or ICs in more than one organ. For details, see Table 2

Evaluation of samples using FISH

A review of histopathology reports during case selection documented visualisation of bacteria in a single case (Table 2, SG13). FISH evaluation of heart, liver and kidney samples from the 23 SG cases demonstrated bacteria in one or more of the selected tissues in 12 (52.2%) cases (95% CI 30.6–73.2). Examples of these can be seen in Figures 2 –4. Of the 23 cases, 10 (43.5%) had bacteria present in the liver (95% CI 23.2–65.5), eight (34.8%) had bacteria in the heart (95% CI 16.4–57.3) and five (21.7%) had bacteria in the kidney (95% CI 7.5–43.7) (Table 1).

Table 2

Case details highlighting age, breed, sex, test results and organs affected

Case ID	Age (years/months)	Breed/sex	Bacteria present (Y/N)	Organ(s)	IC present (Y/N)	Organ(s)	IC type
Study group
SG01	8 years	Sia/M (C)	N	N/A	Y	Lv; Kd	LP
SG02	4 years	DSH/M (C)	N	N/A	N	N/A	N/A
SG03	7 years	DSH/M	N	N/A	Y	Kd	LP
SG04	1 year 1 month (13 months)	DSH/M (C)	Y	Ht, Lv	Y	Lv, Kd	LP (Lv) + Neu (Kd)
SG05	6 months	DSH/F	N	N/A	Y	Ht	LP
SG06	13 years	DSH/M (C)	Y	Ht	Y	Lv, Kd	LP
SG07	7 weeks (1 month 3 weeks)	DSH/M	Y	Ht, Kd	Y	Kd	Hmg
SG08	6 months	DSH/F	N	N/A	Y	Lv	LP
SG09	10 years	DSH/F (S)	Y	Lv	Y	Lv	Neu
SG10	14 years	DLH/M (C)	N	N/A	Y	Lv, Kd	LP
SG11	1 year 2 months (14 months)	DSH/M	N	N/A	Y	Lv	Neu + P
SG12	Unknown	DSH/M	Y	Ht, Lv	Y	Ht, Lv, Kd	LP
SG13	1 week	DSH/M	Y	Lv, Kd	Y	Lv	Nec
SG14	3 months 1 week (13 weeks)	DSH/M (C)	Y	Lv	Y	Lv	LP
SG15	3 weeks	DSH/M	N	N/A	N	N/A	N/A
SG16	2 years	DSH/F (S)	Y	Ht, Lv	N	N/A	N/A
SG17	6 weeks (1 month 2 weeks)	DSH/F	N	N/A	N	N/A	N/A
SG18	10 years	DSH/M (C)	N	N/A	Y	Lv, Kd	LP
SG19	16+ years	DSH/F (S)	N	N/A	N	N/A	N/A
SG20	~10 years	DSH/M (C)	Y	Ht, Lv, Kd	Y	Lv, Kd	LP
SG21	1 year	DSH/M	Y	Lv	Y	Kd	LP
SG22	2 years	DSH/M (C)	Y	Ht, Lv, Kd	Y	Lv, Kd	LP (Kd) + Nec (Lv)
SG23	16 years	DSH/M (C)	Y	Ht, Lv, Kd	Y	Kd	LP
Control group
CG01	6–16 weeks	DSH/M	Y	Ht	N	N/A	N/A
CG02	6+ months	DSH/M	N	N/A	N	N/A	N/A
CG03	5 days	DSH/F	Y	Ht, Kd	N	N/A	N/A
CG04	2 years	DSH/F (S)	N	N/A	Y	LP	Lv
CG05	1–2 years	DSH/M	N	N/A	N	N/A	N/A
CG06	3 months	DSH/F	N	N/A	N	N/A	N/A

Rows in bold = excluded cases (inflammatory cell [IC]-negative cases within the study group; IC-positive cases within the control group)

C = castrated; DLH = domestic longhair; DSH = domestic shorthair; F = female; Hmg = haemorrhage; Ht = heart; Kd = kidney; LP = lymphoplasmacytic; Lv = liver; M = male; Neu = neutrophilic; N/A = none affected; Nec = necrosis; P = plasmacytic; Sia = Siamese; S = spayed

Figure 2

(a–d) Fluorescent in situ hybridisation performed on a section of cat heart using a EUB338 Cy3 labelled probe showing filamentous bacterium (present throughout the sample) within a blood vessel in ventricular cavity, possibly resulting from post-mortem contamination. Diagnosed with pericardial effusion, nephrosis and hepatic vacuolar degeneration. (a) U-MWIG2 (green excitation) filter at × 100. The organism is indicated by the yellow marker. (b) U-MNB2 (blue excitation) filter at × 100. The organism is no longer identifiable because of excitation wavelength change (indicative of successful hybridisation); the associated area of occupation is indicated by the yellow marker. (c) U-MNU2 (UV excitation) filter at × 100. The organism is visible and indicated by the yellow marker. (d) Composite image generated using Adobe Photoshop from red, green and blue images. The organism is clearly identifiable in the red channel and is indicated by the yellow marker. (e) Haematoxylin and eosin (H&E) sample displayed for comparison. H&E staining (performed on a separate slide) showing the same region of feline heart. Note the lack of inflammatory cells and that the organism did not stain. The associated area of occupation is indicated by the yellow marker. RBC = red blood cell

Figure 3

(a–d) Fluorescent in situ hybridisation performed on a section of cat liver (parenchyma) using a EUB338 Cy3 labelled probe. Diagnosed with sepsis, hepatic necrosis and nephritis. (a) U-MWIG2 (green excitation) filter at × 100. A cluster of organisms is encircled in yellow. Individual bacteria are also visible throughout the sample (not marked). (b) U-MNB2 (blue excitation) filter at × 100. The organisms are no longer identifiable because of excitation wavelength change (indicative of successful hybridisation); the associated area occupied by the cluster is encircled in yellow. (c) U-MNU2 (UV excitation) filter at × 100. The organisms are not identifiable – often they are visible within this frequency; the associated area occupied by the cluster is encircled in yellow. (d) Composite image generated using Adobe Photoshop from red, green and blue images. The organisms are clearly identifiable in the red channel; the area occupied by the cluster is encircled in yellow. (e) Haematoxylin and eosin (H&E) sample displayed for comparison. H&E staining (performed on a separate slide) showing the same region of feline liver. Note the lack of inflammatory cells and that the organisms did not stain. The associated area occupied by the cluster is encircled in yellow. H = hepatocyte; RBC = red blood cell; SC = sinusoid capillary

Figure 4

(a–d) Fluorescent in situ hybridisation performed on a section of cat kidney (renal medulla) using a EUB338 Cy3 labelled probe. Diagnosed with sepsis, hepatic necrosis and nephritis. (a) U-MWIG2 (green excitation) filter at × 100. A cluster of organisms is encircled in yellow. (b) U-MNB2 (blue excitation) filter at × 100. The organisms are no longer identifiable because of excitation wavelength change (indicative of successful hybridisation); the associated area occupied by the cluster is encircled in yellow. (c) U-MNU2 (UV excitation) filter at × 100. Again, the organisms are not identifiable within this light frequency; the associated area occupied by the cluster is encircled in yellow. (d) Composite image generated using Adobe Photoshop from the red, green and blue images. The organisms are clearly identifiable in the red channel; again, the area occupied by the cluster is encircled in yellow. (e) H&E sample displayed for comparison. Haematoxylin and eosin (H&E) staining (performed on a separate slide) showing the same region of kidney. Again, note the lack of inflammatory cells and that the organisms did not stain. The associated area occupied by cluster is encircled in yellow. CT = collecting tubule; LH = loop of Henle; VR = vasa recta

Notably, 2/6 (33%) CG cases were found to be positive for bacteria (95% CI 4.3–77.7) when tested with FISH: one (16.7%) case showing organisms in the heart alone (95% CI 0.4–64.1) and one (16.7%) in the heart and kidney (95% CI 0.4–64.1).

All BCS (28/28, 100%) returned positive for bacteria (95% CI 87.7–100.0), demonstrating that the FISH process had been successful. Furthermore, none of the SG or CG slides that tested positive for bacteria returned a positive result when non-eubacterial probes were applied, supporting test specificity. The BCS were also included in non-eubacterial tests, none of which returned a positive result.

Evaluation of samples using H&E staining and comparison with FISH results

Slides collected for H&E staining were used to evaluate all samples in both groups. In total, 18/23 (78.3%) SG cases (95% CI 56.3–92.5) and 1/6 (16.7%) CG case (95% CI 0.4–64.1) showed signs of inflammation in one or more of the tissues (Table 2). Removing SG cases where no ICs were identified and CG cases where ICs were identified, 18 SG cases and five CG cases remained. Of them, 11/18 (61.1%) cats within the SG demonstrated the presence of bacteria on FISH (95% CI 35.7–82.7). Of those, 9/11 (81.8%) demonstrated the presence of ICs within one or more tissues where bacteria were present (95% CI 48.2–97.7) and 2/11 (18.2%) showed ICs in tissues where bacteria were not seen (95% CI 2.3–51.8). Interestingly, within the SG, the presence of bacteria in either the liver or kidney was frequently associated with the presence of ICs – in 7/9 (77.7%) cases (95% CI 40.0–97.2) and 4/5 (80%) cases (95% CI 28.4–99.5), respectively. Furthermore, in all seven (100%) cases where bacteria were identified in the heart, ICs were identified in the kidney (lymphoplasmacytic [LP] in five cases, neutrophilic in one case and haemorrhagic in another) (95% CI 59.0–100.0). Similarly, in these seven cases where bacteria were identified within the heart, ICs were identified in the liver in 5/7 (71.4%) cases (LP in four cases and necrosis in one case) (95% CI 29.0–96.3). However, when bacteria and ICs were identified within the same organ, they did not co-locate. Of the five CG cases in which ICs were not identified, 2/5 (40%) were positive for bacteria on FISH (95% CI 5.3–85.3). Pearson’s χ² test was performed on the refined groupings, demonstrating a χ² value of 0.71 (P = 0.40).

Discussion

FISH was utilised successfully to detect bacteria within post-mortem samples of feline heart, liver and kidney. We identified bacteria in at least one organ in 11/18 (61.1%) SG cases (95% CI 35.7–82.7) compared with 2/5 (40%) CG cases (95% CI 5.27–85.4) within the refined groups (Table 2). Among the SG cases, bacteria were identified within the liver in 9/18 (50%) cases (95% CI 26.0–73.9), the kidney in 5/18 (27.8%) cases (95% CI 9.7–53.4) and the heart in 7/18 (38.9%) cases (95% CI 17.3–64.3). Among the CG cases, bacteria were identified in the heart in 2/5 (40%) cases (95% CI 5.3–85.3) and the kidney in 1/5 (20%) cases (95% CI 0.5–71.6). Routine histopathological examination (after autopsy) noted bacteria in one case only, demonstrating a greater detection rate of bacteria using FISH compared with H&E staining. Furthermore, FISH has not previously been successfully utilised for the detection of bacteria within the feline kidney.

Within our study population, there were three cases with neutrophilic inflammation, one of which demonstrated bacteria in the same organ as the infiltrates. Typically, bacterial infection is associated with neutrophilic infiltrates that are rapidly recruited to sites of infection.¹² Interestingly, of the 14 cases with lymphoplasmacytic inflammation, we demonstrated the presence of bacteria in eight. These were within the same organ in six cases (three cases within the liver, two cases within the kidney, and one case within the liver and kidney). In several cases, bacteria were present in one or more organs where ICs were not detected, with ICs present in the organ or organs where bacteria were not seen. For example, of seven cases that demonstrated bacteria in cardiac sections on FISH, only one showed the presence of ICs within the heart (these were LP in origin); of the six remaining cases, all showed ICs within the liver, kidney or both but none in the heart. Furthermore, as previously mentioned, where bacteria were present in the same organ as the ICs, they did not co-locate. There are several factors that may explain these findings: the unique immunological environment of the heart, the nature of the bacterial infection, the response of the hepatic and renal systems to systemic infection, and the possibility of bacterial translocation.^13
–32

The heart displays a degree of immune privilege, whereby immune responses are limited to prevent significant damage to structures such as valves, the conduction system and less vascular regions of the myocardium.^20,23,26 This may contribute to a lower density of immune cells compared with other organs, resulting in a reduced inflammatory response to bacteria in cardiac tissues.^18,25 Furthermore, the heart’s microenvironment (one resulting from complex interactions between cell types within the heart to maintain cardiac homeostasis and adaptation) may influence immune cell behaviour, creating an environment where bacteria can persist in low numbers without triggering a significant inflammatory response.^15,27

Bacterial species, such as Bartonella species, possess virulence factors that can evade detection or elicit a weaker immune response in host tissues such as the endocardium, while provoking a strong inflammatory response in other organs.^19,28 Chronic conditions, such as renal or hepatic disease, may alter the normal immune response, leading to an increased presence of ICs in those organs while expressing a reduced immune response within cardiac tissues.^13,21,24 Interestingly, of the 11/18 (61.1%) cases in which both ICs and bacteria were present (95% CI 35.7–82.7), 9/11 (81.8%) cases were reported to have had some form of renal and/or hepatic disease (95% CI 48.2–97.7). Of those, 6/9 (66.7%) cases demonstrated bacteria in the heart (95% CI 29.9–92.5), one (16.7%) of which showed bacteria in the heart alone (95% CI 0.4–64.1).

With the hepatic and renal systems highly involved in systemic immunity,^22,31,32 it is possible for an immune response to be triggered in one or both organs, despite the site of infection being elsewhere; this is especially true of the liver, which is seen as a central hub for immune modulation.³⁰ It may explain why these tissues showed the presence of ICs most frequently in this study.

Bacterial translocation may also explain the presence of bacteria in one organ while expressing an inflammatory response in another. Translocation occurs when bacteria shift across the intestinal barrier due to its disruption, allowing entry into portal circulation and central circulation thereafter.¹⁶ As a result of cardiac immune privilege, bacteria entering the circulation via the gut–liver axis could cause localised infection – especially of the endocardium – with minimal immune response within cardiac tissues, while causing a significant immune response within the liver (due to entry via the portal circulation).^14,17,29 Similarly, dysbiosis within the gastrointestinal microbiome can lead to the accumulation of uremic toxins that not only exacerbate the disruption of the intestinal barrier but increase systemic and renal inflammation.³³ This may also explain why a significant number of renal samples (n = 12) showed the presence of ICs while fewer demonstrated the presence of bacteria (n = 5).

Unsurprisingly, the liver was the most common tissue to demonstrate positive FISH results (n = 9), with inflammatory changes being present in seven of these cases. However, areas of inflammation did not align with the apparent infection or colonisation. This dissociation could be attributed to the previously described dynamics occurring at an intra-organ level. However, for this dissociation to occur within a specific tissue, tissue compartmentalisation – the division of organ tissues into subregions by membranes, as well as variance in the composition or structure related to specific regional function³⁴ – should be considered as a possible cause or co-factor. Given the highly compartmentalised structure of the cardiac, hepatic and renal systems,^35
–38 and that the immune response can be isolated within those compartments, it could be hypothesised that inflammation may occur in one area of a tissue while allowing pathogens to exist in another without a corresponding inflammatory response. This is especially pertinent when considering the potential effects of bacterial migration from areas where they are considered commensal to tissues where they are not.^39
–42 Compartmentalisation may also lead to a degree of immune privilege within each organ, which could allow bacteria to inhabit one region with minimal immune response, with an increased response elsewhere.^20,43 It may also create regions of tissue-specific, microenvironmental variance that could mediate the manifestation of infection/inflammation locally within each tissue.^27,32,44,45

Conversely, there may be no correlation between IC and bacteria location within the samples tested, as their occurrence could be unrelated. The sites of inflammation could be due to disease profiles and/or chronic disease of non-pathogenic origin that bear no relationship to infection. For example, hepatic lipidosis causes localised inflammation without pathogenic cause.⁴⁶ Exposure to nephrotoxins or drugs such as non-steroidal anti-inflammatory drugs or aminoglycosides can cause nephritis.⁴⁷ However, cardiac inflammation in cats is often associated with infection;⁴⁸ although it has been suggested that there could be a non-infectious cause, such as toxin exposure, genetic factors or ischemia, there is a paucity of feline studies that address these theories.⁴⁹

Alternatively, bacterial presence with a lack of inflammatory response may be due to early stages of infection, especially where there is immunocompromise. Studies on feral cat populations in Saint Kitts demonstrated a prevalence of feline immunodeficiency virus (FIV) in the range of 18–36%, largely within adult male populations. Of the owned feline population in our study, none were found to be FIV positive.^50,51 Given the high proportion of male cats aged over 1 year that were IC positive in this study (11/15, 73.3%, 95% CI 44.9–92.2], of which 6/11 (54.5%, 95% CI 23.4–83.3) demonstrated the presence of both bacteria and ICs, FIV may play a confounding role. However, among the IC-positive adult males, only one had been tested for FIV – and was positive (1/11, 9.1%, 95% CI 0.2–41.3) – and no bacteria were found in this case. The remaining 10/11 (90.9%) cats had not been tested (95% CI 58.7–99.8). Furthermore, of the six adult males that demonstrated the presence of bacteria and ICs, five (83.3%) were owned (95% CI 35.9–99.6). This would suggest that FIV did not affect the findings, but it must be noted that feral and domestic feline populations on Saint Kitts regularly interact, and given the lack of FIV testing among the cats, it remains plausible that many of the owned cats in this study were FIV positive.

Limitations

Retrospective studies have inherent issues that limit their reliability. For example, histories may not contain all pertinent information, which can lead to selection bias or limit the numbers included. In turn, this may render a study underpowered or unrepresentative of a population.^52,53 In this study, case selection was based on a search of the records for inflammation and/or diseases related to inflammation. Despite the records of all cases included within the study group noting its presence, five were excluded as no ICs were found. Similarly, cases were recruited into the control group by searching through records for presentation unlikely to be associated with inflammation. However, these cases were of differing post-mortem ages and in various traumatic states. Post-mortem contamination due to the early stages of decomposition, specifically autolysis, which displays patterns of cellular breakdown similar to those of necrosis within a tissue but without the presence of ICs, coupled with the early stages of putrefaction, may account for the presence of bacteria without the presence of ICs in two of the CG cases.^54
–57

Contamination may have also been caused by the ante-mortem introduction of bacteria through the terminal mechanism of injury or by post-mortem bacterial translocation facilitated by said mechanism.⁵⁸ These limitations potentially influenced study power as the assumption was made that the CG cases would not have any bacteria or inflammation present. Considering that 40% of included cases within the CG demonstrated the presence of bacteria, this challenges the assumption that the CG would serve as a viable baseline, introducing a variability that may have obscured real differences between the groups and reducing the detection of statistically significant effects.⁵⁹ Increases in the within-group variance would also inflate the standard error and diminish the likelihood of achieving significance unless the sample size was sufficiently large.⁶⁰ As such, a minimum of 87 cases would have been required in both groups to demonstrate a significant difference (with a power of 0.8). When samples were selected for inclusion, the RUSVM pathology database did not have this number of cases matching the inclusion criteria. This emphasises the broader challenge in studies of this nature: that of the difficulty of assembling sufficiently large and well-matched cohorts.⁶¹ The findings from the control group not only impact the interpretation of results but also highlight the need for more robust methodological frameworks in such studies.⁶¹

Furthermore, this study was conducted on a small island with a degree of uniqueness within its feline population, due to genetic restriction and the prevalence of certain disease types.⁶² These factors limit the generalisability of the results. In addition, the small sample sizes increase the likelihood of random error or type II errors, which, despite all efforts to minimise them, may reduce the reliability of the findings.^63,64

Future studies should be prospective in nature, incorporating a multicentred approach to allow recruitment of larger cohorts within both the SG and CG, and to increase the generalisability of results. Studies must also incorporate the use of species-specific FISH probes alongside techniques such as PCR/16s rRNA gene sequencing, preferably on fresh tissue samples, better allowing for the molecular characterisation of bacteria.⁸ Although deemed beyond the scope of this small pilot study, future studies must also incorporate quantitative analysis and detailed descriptive patterns to accurately report any bacteria-inflammation spatial relationships.

Conclusions

Despite the preliminary nature of the findings, the small sample size and the retrospective design of this study, FISH successfully detected bacteria within archival FFPE samples of feline heart, liver and kidney. This is the first study that has identified bacteria within the feline kidney using FISH. Only a single case documented the visualisation of bacteria using H&E staining, whereas an additional 11 cases were identified using FISH. This demonstrates that FISH is an excellent complementary diagnostic tool when available and suggests that greater vigilance is required for the presence of bacteria. However, given the lack of power and other limitations noted during this research, further studies would need to be undertaken to demonstrate any such conclusion or before offering any direct clinical recommendations. Any such studies must be prospective in nature and incorporate multiple molecular techniques on fresh tissue samples to accurately speciate any bacteria found.

Supplemental Material

Clinically confirmed or suspected disease processes (SG cases) prior to H&E review.

Footnotes

Acknowledgements

The authors would like to express their thanks to Marianne Rivers (research volunteer; DVM student) for her assistance with the sample selection and assessment before study inclusion.

Accepted: 3 October 2025

Supplementary material

The following file is available as supplementary material:

Clinically confirmed or suspected disease processes (SG cases) prior to H&E review.

Conflict of interest

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

This research is part of a project funded by the Center for Integrative Mammalian Research and the Center for One-Health, Zoonosis, and Tropical Medicine, Ross University School of Veterinary Medicine (grant number 41008-2022).

Ethical approval

The work described in this manuscript involved the use of non-experimental (owned or unowned) animals. Established internationally recognised high standards (‘best practice’) of veterinary clinical care for the individual patient were always followed and/or this work involved the use of cadavers. Ethical approval from a committee was therefore not specifically required for publication in JFMS. Although not required, where ethical approval was still obtained, it is stated in the manuscript.

Informed consent

Informed consent (verbal or written) was obtained from the owner or legal custodian of all animal(s) described in this work (experimental or non-experimental animals, including cadavers, tissues and samples) for all procedure(s) undertaken (prospective or retrospective studies). For any animals or people individually identifiable within this publication, informed consent (verbal or written) for their use in the publication was obtained from the people involved.

ORCID iD

Matthew J Rolph

Pompei Bolfa

Kerry E Rolph

References

Almeida

Azevedo

NF.

An introduction to fluorescence in situ hybridization in microorganisms. In: Azevedo

Almeida

(eds). Fluorescence in-situ hybridization (FISH) for microbial cells. New York: Humana, 2021, pp 1–15.

Popescu

FISHing for RNA: an overview of fluorescent in situ hybridization methods for RNA detection.

https://blog.biodock.ai/overview-of-rna-fish/ (2021, accessed 11 November 2023).

Prudent

Raoult

Fluorescence in situ hybridization, a complementary molecular tool for the clinical diagnosis of infectious diseases by intracellular and fastidious bacteria.

FEMS Microbiol Rev 2019; 43: 88–107.

Young

Jackson

Wyeth

RC.

A technical review and guide to RNA fluorescence in situ hybridization.

PeerJ 2020; 8. DOI: 10.7717/peerj.8806.

Moter

Gobel

UB.

Fluorescence in situ hybridization (FISH) for direct visualization of microorganisms.

J Microbiol Methods 2000; 41: 85–112.

Lukumbuzya

Schmid

Pjevac

, et al. A multicolor fluorescence in situ hybridization approach using an extended set of fluorophores to visualize microorganisms. Front Microbiol 2019; 10. DOI: 10.3389/fmicb.2019.01383.

Simpson

Dogan

Rishniw

, et al. Adherent and invasive Escherichia coli is associated with granulomatous colitis in Boxer dogs. Infect Immun 2006; 74: 4778–4792.

Rolph

Bolfa

Cavanaugh

, et al. Fluorescent in situ hybridization for the detection of intracellular bacteria in companion animals. Vet Sci 2024; 11 . DOI: 10.3390/vetsci11010052.

Neugent

Gadhvi

Palmer

, et al. Detection of tissue-resident bacteria in bladder biopsies by 16S rRNA fluorescence in situ hybridization. J Vis Exp 2019; 152. DOI: 10.3791/60458.

10.

Sigma-Aldrich . User guide – eubacteria fluorescent in situ hybridization (FISH) probe for bacterial detection. Darmstadt, Germany: Merk, 2021, p 8.

11.

Cambio. Fluorescence in situ hybridisation protocol for whole chromosome paints. Cambridge: Cambio, 2009, p 4.

12.

Hostetter

SJ.

Neutrophil function in small animals.

Vet Clin North Am Small Anim Pract 2012; 42: 157–171.

13.

Álvarez-Fernández

Breitschwerdt

Solano-Gallego

Bartonella infections in cats and dogs including zoonotic aspects.

Parasit Vectors 2018; 11 . DOI: 10.1186/s13071-018-3152-6.

14.

Dinakaran

Microbial translocation in the pathogenesis of cardiovascular diseases: a microbiome perspective.

J Cardiol Curr Res 2017; 8. DOI: 10.15406/jccr.2017.08.00305.

15.

Frangogiannis

NG.

Regulation of the inflammatory response in cardiac repair.

Circ Res 2012; 110: 159–173.

16.

Fukui

Gut–liver axis in liver cirrhosis: how to manage leaky gut and endotoxemia.

World J Hepatol 2015; 7: 425–442.

17.

Fukui

Role of gut dysbiosis in liver diseases: what have we learned so far?

Diseases 2019; 7. DOI: 10.3390/diseases7040058.

18.

Gan

Jin

Ren

The immune map of human body.

bioRxiv [Express]. Posted online 6 March 2024. DOI: 10.1101/2024.03.03.583227.

19.

Harms

Dehio

Intruders below the radar: molecular pathogenesis of Bartonella spp.

Clin Microbiol Rev 2012; 25: 42–78.

20.

Hill

Kwon

Gerry

, et al. Immune privilege of heart valves. Front Immunol 2021; 12 . DOI: 10.3389/fimmu.2021.731361.

21.

Khanam

Kottilil

Abnormal innate immunity in acute-on-chronic liver failure: immunotargets for therapeutics.

Front Immunol 2020; 11 . DOI: 10.3389/fimmu.2020.02013.

22.

Kurts

Panzer

Anders

, et al. The immune system and kidney disease: basic concepts and clinical implications. Nat Rev Immunol 2013; 13: 738–753.

23.

Mann

The emerging role of innate immunity in the heart and vascular system: for whom the cell tolls.

Circ Res 2011; 108: 1133–1145.

24.

Raybould

Morales

, et al. Bartonella endocarditis and pauci-immune glomerulonephritis: a case report and review of the literature. Infect Dis Clin Pract (Baltim Md) 2016; 24: 254–260.

25.

Sender

Weiss

Navon

, et al. The total mass, number, and distribution of immune cells in the human body. Proc Natl Acad Sci U S A 2023; 120. DOI: 10.1073/pnas.2308511120.

26.

Swirski

Nahrendorf

Cardioimmunology: the immune system in cardiac homeostasis and disease.

Nat Rev Immunol 2018; 18: 733–744.

27.

Tzahor

Dimmeler

A coalition to heal – the impact of the cardiac microenvironment.

Science 2022; 377. DOI: 10.1126/science.abm4443.

28.

Liu

, et al. Sneaky tactics: ingenious immune evasion mechanisms of Bartonella. Virulence 2024; 15. DOI: 10.1080/21505594.2024.2322961.

29.

Wang

Feng

, et al. The gut microbiota and its interactions with cardiovascular disease. Microb Biotechnol 2020; 13: 637–656.

30.

Yan

The role of the liver in sepsis.

Int Rev Immunol 2014; 33: 498–510.

31.

Zhang

Extrahepatic factors in hepatic immune regulation.

Front Immunol 2022; 16. DOI: 10.3389/fimmu.2022.941721.

32.

Zhao

Huang

, et al. Current insights into the hepatic microenvironment and advances in immunotherapy for hepatocellular carcinoma. Front Immunol 2023; 18. DOI: 10.3389/fimmu.2023.1188277.

33.

Summers

Quimby

Insights into the gut-kidney axis and implications for chronic kidney disease management in cats and dogs.

Vet J 2024; 306. DOI: 10.1016/j.tvjl.2024.106181.

34.

van Gelder

. Principles of compartmentation. In: Lajtha

(ed). Experimental neurochemistry. Boston, MA: Springer, 1982, pp 183–206.

35.

Arackal

Alsayouri

Histology, heart. In: StatPearls. Treasure Island, FL: StatPearls Publishing, 2023.

36.

Balzer

Rohacs

Susztak

How many cell types are in the kidney and what do they do?

Annu Rev Physiol 2022; 84: 507–531.

37.

Mahadevan

Anatomy of the liver.

Surgery (Oxf) 2020; 38: 427–431.

38.

Vekemans

Braet

Structural and functional aspects of the liver and liver sinusoidal cells in relation to colon carcinoma metastasis.

World J Gastroenterol 2005; 11: 5095–5102.

39.

Belkaid

Naik

Compartmentalized and systemic control of tissue immunity by commensals.

Nat Immunol 2013; 14: 646–653.

40.

Chen

Ishigami

Doi

, et al. The types and proportions of commensal microbiota have a predictive value in coronary heart disease. J Clin Med 2021; 10. DOI: 10.3390/jcm10143120.

41.

Lang

Schnabl

Microbiota and fatty liver disease – the known, the unknown, and the future.

Cell Host Microbe 2020; 28: 233–244.

42.

Liu

Zhang

, et al. Kidney microbiota dysbiosis contributes to the development of hypertension. Gut Microbes 2022; 14. DOI: 10.1080/19490976.2022.2143220.

43.

Tian

The liver works as a school to educate regulatory immune cells.

Cell Mol Immunol 2013; 10: 292–302.

44.

Anders

Ryu

Renal microenvironments and macrophage phenotypes determine progression or resolution of renal inflammation and fibrosis.

Kidney Int 2011; 80: 915–925.

45.

Litvinˇuková

Talavera-López

Maatz

, et al. Cells of the adult human heart. Nature 2020; 588: 466–472.

46.

Dimski

Taboada

Feline idiopathic hepatic lipidosis.

Vet Clin North Am Small Anim Pract 1995; 25: 357–373.

47.

Monaghan

Nolan

Labato

Feline acute kidney injury: 1. Pathophysiology, etiology and etiology-specific management considerations.

J Feline Med Surg 2012; 14: 775–784.

48.

Ferasin

Feline myocardial disease. 1: Classification, pathophysiology and clinical presentation.

J Feline Med Surg 2009; 11: 3–13.

49.

Fox

PR.

Endomyocardial fibrosis and restrictive cardiomyopathy: pathological and clinical features.

J Vet Cardiol 2004; 6: 25–31.

50.

Chi

Fang

Koster

, et al. Prevalence of feline immunodeficiency virus and Toxoplasma gondii in feral cats on St. Kitts, West Indies. Vet Sci 2021; 8. DOI: 10.3390/vetsci8020016.

51.

Kelly

Stocking

Gao

, et al. Identification of feline immunodeficiency virus subtype-B on St. Kitts, West Indies by quantitative PCR. J Infect Dev Ctries 2011; 5: 480–483.

52.

Nickson

Retrospective studies and chart reviews.

https://litfl.com/retrospective-studies-and-chart-reviews/ (2020, accessed 23 December 2024).

53.

Talari

Goyal

Retrospective studies – utility and caveats.

J R Coll Physicians Edinb 2020; 50: 398–402.

54.

Almulhim

Menezes

RG.

Evaluation of postmortem changes. In: StatPearls. Treasure Island, FL: StatPearls Publishing, 2023.

55.

Brooks

JW.

Postmortem changes in animal carcasses and estimation of the postmortem interval.

Vet Pathol 2016; 53: 929–940.

56.

Shedge

Krishan

Warrier

, et al. Postmortem changes. In: StatPearls. Treasure Island, FL: StatPearls Publishing, 2023.

57.

Wenzlow

Neal

Stern

, et al. Feasibility of using tissue autolysis to estimate the postmortem interval in horses. J Vet Diagn Invest 2021; 33: 825–833.

58.

Pokryshko

Khomenko

Experimental translocation of intestinal bacteria caused by closed abdominal trauma, acute blood loss, internal haemorrhage.

Bull Med Biol Res 2023; 16: 15–22.

59.

Magill

Knight

McCrone

, et al. A scoping review of the problems and solutions associated with contamination in trials of complex interventions in mental health. BMC Med Res Methodol 2019; 19. DOI: 10.1186/s12874-018-0646-z.

60.

Biau

Kernéis

Porcher

Statistics in brief: the importance of sample size in the planning and interpretation of medical research.

2008; 466: 2282–2288.

61.

Drude

Martinez-Gamboa

Danziger

, et al. Planning preclinical confirmatory multicenter trials to strengthen translation from basic to clinical research – a multi-stakeholder workshop report. Transl Med Commun 2022; 7. DOI: 10.1186/s41231-022-00130-8.

62.

Kristiansson

Naukkarinen

Peltonen

Isolated populations and complex disease gene identification.

Genome Biol 2008; 9. DOI: 10.1186/gb-2008-9-8-109.

63.

Knudson

Lindsey

Type I and Type II errors in correlations of various sample sizes.

Compr Psychol 2014; 3: 1–5.

64.

Thiese

Ronna

Ott

P value interpretations and considerations.

J Thorac Dis 2016; 8: E928–E931.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.02 MB