Abstract
Tissues from 34 naturally feline immunodeficiency virus (FIV)-infected cats, 13 asymptomatic cats and 21 cats with signs of feline acquired immunodeficiency syndrome (F-AIDS), and 35 FIV-seronegative subjects were examined to determine the presence of amyloid deposits. Twenty experimentally FIV-infected cats and five specific pathogen-free (SPF) control cats were also included in the study. Paraffin-embedded sections from kidney and other organs were submitted to histological and histochemical analysis. Amyloid deposits were identified by a modified Congo red stain and confirmed by electron microscopy to demonstrate the presence of amyloid fibrils in amyloid positive glomeruli. In all positive cases, secondary amyloidosis was identified with potassium permanganate pretreatment and amyloid type was further characterised by immunohistochemistry using primary antibodies against human AA and feline AL amyloids. Amyloid deposits were present in different tissues of 12/34 (35%) naturally FIV-infected cats (seven presenting F-AIDS and five in asymptomatic phase) and in 1/30 FIV-seronegative cats. All the experimentally FIV-infected and SPF subjects showed no amyloid deposits. Amyloidosis has been reported in human lentiviral infections, and the data reported here demonstrate the need, in naturally FIV-infected cats, to consider the presence of amyloidosis in differential diagnosis of hepatic and renal disorders to better assess the prognosis of the disease.
Introduction
The generic name of amyloidosis is applied to a group of diseases caused by extracellular deposition in various tissues throughout the body of approximately 10 nm insoluble fibrillar proteins with a β-pleated sheet conformation. Amyloidosis is associated with a heterogeneous group of disorders and it may be classified by the chemical nature of the protein fibrils, which constitute the amyloid deposit. Secondary amyloidosis associated with states of chronic infection (AA amyloid) and amyloidosis related to plasma cell dyscrasia (AL amyloid) are of major importance: 1 AA amyloid is derived from a characteristic acute phase protein, serum amyloid protein A (SAA), complexed mainly by a high density lipoprotein in serum, and characteristic of tissue damage and inflammation; 2 AL amyloid is derived from light chain immunoglobulins.
The association of amyloidosis with retroviral infections has been reported previously in lentiviral infections, particularly in human deficiency virus (HIV)- infected patients,3 –5 but little is known about the possible correlation between feline immunodeficiency virus (FIV) infection and amyloidosis. The presence of amyloid deposits has been reported sporadically in naturally infected cats,6,7 and the nature of amyloid deposits has never been studied. The aim of this study was to investigate the presence and distribution of amyloidosis in naturally and experimentally FIV-infected cats and controls, and to characterise the type of amyloid deposits detected.
Materials and methods
Thirty-four naturally FIV-infected cats and 30 FIV-seronegative cats, matched as closely as possible for age, sex, breed and lifestyle, were identified among the cats submitted for necropsy at the Department of Animal Pathology of the School of Veterinary Medicine of the University of Pisa. FIV infection was established by detecting antibodies to FIV using a commercial kit (Cite Combo FIV-FeLV; Agritech Systems) and an indirect fluorescent antibody assay using FIV-infected Crandel feline kidney cells as substrate. Positive results were confirmed by Western blot analysis on electrophoresed FIV proteins from virus cultured onto FL4 cells. 8 All the FIV-seropositive cats included in this study were negative for feline leukaemia virus (FeLV) antigen, as demonstrated by a commercial ELISA kit (Cite Combo FIV-FeLV; Agritech Systems) and feline infectious peritonitis antibodies (Diasystems Celisa FIP; Tech America).
At the time of sampling, 13 of the FIV-infected cats were asymptomatic, 11 presented signs compatible with the diagnosis of feline AIDS-related complex (recurrent fever of undetermined origin, lymphoadenopathy and weight loss) and 10 cats showed clinical signs of chronic opportunistic infections suggestive of feline-acquired immunodeficiency syndrome (F-AIDS). 9
Of the FIV-seropositive cats 25 were male and nine were female, and were aged between 3.5 and 13 years (mean 7.5 years ± 1.8 years), while of the FIV-seronegative cats 19 were male and 11 were female, and aged between 3 and 13 years (mean 7.6 years ± 2.0 years). Twenty specific pathogen-free (SPF) cats purchased from Iffa Credo (L’Arbresle) and infected with the local Pisa-M2 FIV isolate at 20–30 weeks of age were also examined. Cats were maintained in accordance with the Ethical Committee guidelines, in an isolation unit of the Department of Biomedicine of the University of Pisa (Techniplast, Buguggiate, VA, Italy) throughout the period of observation. Another group of five SPF cats, maintained in the same environment as uninfected controls, was also studied. The animals were fed a standard diet (Hendrix, Verona, Italy) and they received full clinical examination and routine haematological and biochemical tests at regular intervals. T-cell subsets were examined by flow cytometric analysis as described. 10 Flow cytometry analysis was performed using fluorescein-conjugated murine monoclonal antibodies to feline CD4 and CD8 T-cells surface markers (Southern Biotech) and an Epics Elite cell analyser (Coulter Electronics). The infection with Pisa-M2 FIV isolate leads to seroconversion in 4–6 weeks and to a reduction of circulating CD4+ lymphocytes by almost two-thirds in 1 year. From 6 months to 2 years post-infection all the FIV-infected and control cats were sedated with ketamine hydrochloride (50 mg IM), killed humanely by intravenous injection of Tanax (Hoechst AG) and submitted immediately to necropsy.
Samples obtained from kidneys, heart, skin, lung, stomach, small intestine, liver, spleen, pancreas, urinary bladder, joints, lymph nodes (cervicalis superficialis, popliteal, and mesenterialis), bone marrow, thyroid, parathyroid, adrenal gland, brain, spinal cord and eyes were collected in 10% buffered formalin (pH 7.4) and processed by routine methods for histological examination. Sections (4 μm thick) were stained with haematoxylin and eosin for light microscopic examination. Various inflammatory lesions, when detected macroscopically or microscopically, were stained with Ziehl-Neelsen acid-fast, Gram and periodic acid-Schiff (PAS) stain to evaluate their origin. Kidneys were stained with phosphotungstic acid–haematoxylin and PAS methenamine silver (PASM) stains. Amyloid was demonstrated by alkaline Congo red staining with polarisation on 8 μm sections. 11 The amount of amyloid deposits in Congo red-stained tissue sections was scored as 0 = negative; + = focal; ++ = diffuse, as reported previously. 6 Differentiation between primary and secondary amyloidosis was performed with modified Congo red stain with potassium permanganate pretreatment. 12 The type of fibrils in amyloid deposits was further characterised by immunohistochemistry using antisera cross-reacting with feline amyloid proteins. 13
After de-paraffinisation, immunohistochemistry (IHC) was performed using the biotin–streptavidin–peroxidase method. Endogenous peroxidases were blocked by incubation in 0.5% hydrogen peroxide in methanol for 10 mins. The primary antibodies used (kindly gifted by RP Linke) were an anti-AA monoclonal murine antibody (MoAb mc4 against human AA protein, culture supernatant) at a dilution of 1:30, a rabbit polyclonal anti-human AA at a dilution of 1:50 and a rabbit polyclonal anti-cat AL at a dilution of 1:250. After serum blocking, sections were incubated with the primary antibodies at 4°C overnight. The sections were then incubated sequentially at room temperature for 30 mins with a pan-specific biotinylated secondary antibody (Vectastain; Vector Laboratories) diluted in Tris buffer saline solution pH 8.4, as suggested by manufacturer’s instructions. After three washes, sections were treated for 1 h with peroxidase-labeled streptavidin (Biogenex Laboratories) diluted 1:100 in Tris buffer. The chromogen used was 3,3’-diaminobenzidine tetrahydrocloride, and then the sections were counterstained with Gill’s haematoxylin. As controls the primary antibodies were replaced with non-immune rabbit serum or replacing the primary monoclonal antibodies with a murine subclass matched unrelated antibodies. Sections of kidney from a cat with a previously diagnosed renal amyloidosis were used as positive controls. For IHC analysis the scoring system was based on the immunostaining intensity as follows: weak (+); moderate (++); strong (+++).
For electron microscopy ~1 mm3 pieces of selected specimens (renal cortex, liver and spleen) from amyloid-positive cats were fixed in Karnovsky’s solution, rinsed overnight with phosphate-buffered saline solution (PBSS) pH 7.3, post-fixed in omium tetroxide 1% in PBSS (pH 7.3) and embedded in Epon. Survey sections with glomeruli and other areas of interest were cut at 0.5 mm and stained with toluidine blue. Ulthathin sections, double stained with 2% uranyl acetate and lead citrate, from a total of 3–4 representative blocks per case were cut and examined with a Philips CM10 electron microscope.
Statistical analysis was performed using the statistical package SPSS Advanced Statistics 13.0 (SPSS). A χ2 test was used to investigate the significance of the relationship between individual variables in the different groups of subjects examined. Statistical significance was based on a 5% (0.05) significance level.
Results
The clinical findings of the 34 naturally FIV-infected cats and of the 30 FIV-uninfected subjects are presented in Table 1. The presence of haematological abnormalities, chronic infection of the oral cavity and of the upper respiratory tract, and the presence of proteinuria was statistically higher in FIV-infected cats (P <0.001), while the presence of tumours, peritonitis and pleuritis were statistically higher in FIV-uninfected cats.
Clinical findings in 34 naturally FIV-infected and 30 FIV-seronegative cats
The main pathological findings observed at necropsy in naturally FIV-infected cats were lymphadenomegaly (23/34), renal alterations (22/34), upper and lower respiratory tract inflammation (19/34), bad nutritional conditions (14/34), ulcerative stomatitis (13/34) and chronic enteritis (10/34). In FIV-seronegative subjects pathological findings were markedly less frequent: 11/30 cats presented macroscopic signs of peritonitis and 6/30 showed signs of pleuritis due to FIP infection. Renal alterations (3/30), ulcerative stomatisis (2/30), upper and lower respiratory tract inflammation (2/30) and skin lesions (1/30) were also detected. Lymphadenomegaly, renal alterations, respiratory tract inflammation and stomatitis were more frequent in FIV-infected cats than in uninfected subjects (P <0.001).
In the group of 20 SPF experimentally FIV-infected cats, the main necroscopic lesions were lymphadenomegaly (12/20), upper and lower respiratory tract inflammation (10/20), chronic enteritis (9/20), renal alterations (8/20) and ulcerative stomatitis (4/20). No alterations were observed in the group of five SPF cats.
Microscopic examinations revealed the presence of diffuse amyloidosis in 12/34 (35%) naturally FIV-infected subjects and in 1/30 (3%) FIV-seronegative control animals.
Amyloid was located mainly in kidneys, with large amounts in glomeruli and renal interstitium (Figure 1a, b). The spleen and adrenals were not enlarged, but, histologically, masses of amyloid were seen in the red pulp of the spleen and in the cortex of the adrenal glands. Foci of amyloid were seen in the liver (Disse’s spaces and walls of vessels; Figure 1c), submucosa of the intestinal tract, and the walls of blood vessels of the gut, pancreas, salivary gland and lymph nodes. The distribution of amyloidosis in the organs of naturally FIV-infected subjects is reported in Table 2. Kidneys were affected in all the 12 naturally FIV-infected cats, while spleen and liver were affected in 11 subjects. Heart, central nervous system, stomach, lung and trachea never presented amyloid deposits. In 6/12 cases amyloid deposits were associated with plasmacytosis.
Amyloid deposits in naturally FIV-infected cats. (a) Kidney, glomerular amyloidosis. Diffuse increase of capillary walls due to amyloidosis deposition — Congo red stain (bar = 80 μm). (b) Kidney, glomerular amyloidosis. Very strong immunohistochemical cross-reaction of amyloid deposits using anti-human AA monoclonal antibody (mc4). Biotin–streptavidin–peroxidase method (bar = 80 μm). (c) Liver, hepatic amyloidosis. AA amyloid deposits in the Disse’s space and in the walls of small vessels. Biotin–streptavidin–peroxidase method (bar = 20 μm). (d) Kidney, glomerular amyloidosis. Electron micrograph showing subendothelial deposition of amyloid fibrils (bar = 1 μm)
Amyloid distribution in FIV-infected and FIV-uninfected amyloid-positive cats
CNS = Central nervous system; ND = not detected; – = negative; + = focal deposits distribution; ++ = diffuse deposits distribution
Of the 12 naturally FIV-infected cats presenting amyloidosis, seven showed F-AIDS signs and five were asymptomatic. Amyloid deposits were never observed in either the 20 experimentally FIV-infected cats or in the five SPF controls maintained in the isolation unit. Using a χ2 test, a significant difference between amyloid presence in naturally FIV-infectetd cats and non-infected cats was noted (P <0.001).
The results of investigations to differentiate amyloid deposits with histochemical and immunohistochemical investigations are summarised in Table 3. In all the 13 positive cats, histochemistry and immunohistochemistry showed that the deposits were composed by secondary amyloid (Figure 1d).
Histochemical and immunohistochemical staining of amyloid deposits observed in FIV-infected and FIV-uninfected amyloid-positive cats
IHC = immunohistochemistry; MM = mouse monoclonal; RP = rabbit polyclonal; – = negative; + = weak immunostaining intensity; ++ = moderate immunostaining intensity; +++ = strong immunostaining intensity
Amyloid fibrils were detected by electron microscopy in intramembranous or subendothelial regions of the basement membrane (Figure 1d) and in the mesangial areas of glomeruli or renal interstititum.
Ziehl-Neelsen acid-fast and PAS stains in naturally FIV-infected and uninfected cats showing amyloidosis gave negative results. In the same subjects, Gram staining showed the presence, by inflammatory reactions, of non-specific infections in some of the organs involved.
Discussion
Our study allowed us to detect amyloidosis in approximately one-third of naturally FIV-infected cats: amyloid deposits were found in different organs of 12/34 unselected cats infected naturally with FIV, whereas in control FIV-seronegative cats the incidence was rare (one subject of 30). On the contrary, in experimentally FIV-infected subjects amyloid deposits were not detected, confirming a previous extensive study. 7 On the basis of these results the prevalence of amyloidosis in naturally FIV-infected cats was higher than in FIV-uninfected cats. Kidneys, spleen, liver, pancreas and adrenal gland were commonly involved, while deposits in lymph nodes and in the intestinal wall were less frequent. In the FIV-uninfected cat which scored positive, amyloid deposits were detected only in kidneys and spleen.
Histochemical and IHC analysis showed that amyloid deposits detected both in FIV-infected and uninfected cats were always of secondary type. Congo red stain modified with potassium permanganate (KMnO4) pretreatment, according to Wright, 12 is a widely diffused method to differentiate primary (AL) and secondary amyloid (AA). AA-amyloid type is sensitive to KMnO4 and, after this pretreatment, it loses its affinity to Congo red staining. However, this is an indirect histochemical procedure and its results must be confirmed by IHC analysis. 1 For this reason all the Congo red-positive tissues were submitted to IHC procedures. Two antibodies against AA-protein (mouse monoclonal and rabbit polyclonal) and an antibody against AL-protein were used. Amyloid deposits always immunoreacted for AA antibodies, both mouse and rabbit serum. No positivity was observed to AL antibody, confirming the histochemical results. These results confirmed that cross-reactivity of anti-AA antibodies allows diagnosis of AA-type amyloid in species against which no anti-AA reagent is available. In spite of the possible chemical heterogeneity of protein AA among species, and in some cases among individuals of the same species, the epitopes detected by mc4 MoAb is preserved evolutionarily because it can be identified in several animal species.13,14
The association of amyloidosis with retrovirus infections has been reported previously. Amyloidosis has been observed as complication of caprine arthritis-encephalitis infection in goats, even if rarely observed in this species. 15 Significant amounts of amyloid were also observed in the malpighian corpuscles of the spleen, in the pancreatic islets and in the lamina propria of the small intestine of 4/16 macaques that died with AIDS at a primate research centre. 16 Furthermore, in a group of 18 Rhesus monkeys infected experimentally with the Simian immunodeficiency virus/Delta 2, two had significant amounts of amyloid in spleen, lymph nodes and lamina propria of the small intestine. 17 Amyloidosis has also been recognised rarely in the kidney of HIV-infected patients,3–5 even if patients with AIDS and AIDS-related complex presented an elevation of serum amyloid precursor concentrations. 18 Recent studies revealed that an induction of acute phase protein serum amyloid (A-SAA) occurred during the acute phase of HIV-1 infection, 19 and that in patients with advanced HIV-1-infection, increased SAA levels accompanied cytomegalovirus infection. 20
As reported previously, in FIV infection the presence of amyloidosis has been ascertained only in the kidney of naturally-infected subjects.6,7 Our study allowed us to determine that amyloid deposits in FIV-infected cats were due to AA-amyloid deposition, which is associated with chronic inflammation conditions and not with an overproduction of immunoglobulin light chains due to a plasma cell discrasia. The absence of amyloid deposits in experimentally FIV-infected cats maintained in an isolation unit with no opportunity to contract any secondary infection, demonstrated that lentiviral infection alone is not sufficient for the development of amyloidosis, even if the short period of infection (24 months) could be the cause. Chronic FIV infection appeared to modify the acute phase response and Mycoplasma species infections resulted associated with a significant increase in SAA concentrations. 21
In naturally FIV-infected cats, kidney amyloid had both medullary and glomerular location; the latter, in particular, can be considered a negative prognostic factor that can worsen proteinuria and lead to a severe renal failure. Of 16 cats presenting clinical signs of renal damage, only four did not present renal amyloid deposits. This data shows how amyloidosis could be an important element for the onset of these signs. Cats with naturally acquired FIV infection were at increased risk of proteinuria. A recent study demonstrated that the prevalence of proteinuria was higher in naturally FIV-infected cats than in FIV-uninfected cats; however, there was no significant difference in the prevalence of azotaemia. 22 The results of this study were consistent with those of previous investigations23,24 and, in another study, young cats with chronic kidney disease were significantly more likely to be FIV-infected compared with young cats without renal disease, even if FIV infection was not associated with renal disease in older cats. 25 Thus, renal amyloidosis should be an important cause of heavy proteinuria in the feline species and it should, therefore, be interesting to extensively investigate the presence of amyloidosis in naturally FIV-infected subject.
Even hepatic amyloidosis is an important finding in affected subjects. Eleven of 34 naturally FIV-infected cats presented amyloid deposits located in the liver. This location can lead to an aggravation of the animal’s condition, facilitating liver insufficiency, rupture and haemorrhages, as reported previously.26,27 Prognosis can be also worsened by splenic and pancreatic amyloid deposits, detected in 11/34 and 9/34 of our study subjects, respectively, as well by the involvement of lymph nodes, adrenal glands and guts, even if less frequent.
Conclusions
To the best of our knowledge, this is the first study to investigate the presence and distribution of amyloidosis in association with natural FIV infection. The results described herein suggest that amyloidosis can be a frequent finding in FIV-infected cats, particularly in renal tissue. This location could be a further negative prognostic factor that may increase proteinuria and induce severe renal failure. The further presence of amyloid deposits in other organs, such as the liver, can lead to an aggravation of the condition of the affected subjects. For these reasons amyloidosis seems to be an important element for the prognosis of FIV-infected cats.
Footnotes
Addendum
The cats in this paper were experimentally infected with FIV, in the context of a large project for the development of a vaccine for human AIDS (M Pistello, D Matteucci, F Bonci, et al. AIDS vaccination studies using an ex vivo feline immunodeficiency virus model: protection from an intraclade challenge administered systemically or mucosally by an attenuated vaccine. J Virol 2003; 77: 10740–10750). They were euthanased at different times post-infection to verify the viral load and the distribution of the tissue alterations. These histopathological examinations were also used to investigate the pathogenesis of FIV-induced tissue changes and, for the current paper, the presence of amyloid deposits.
Funding
This work was partially supported by the Ministry of Health - Istituto Superiore di Sanità, 4th AIDS National Research Program, ‘AIDS National Project’ (Italian Concerted for AIDS Vaccination).