**4. Discussion**

The results of our investigation clearly demonstrate that FIV infection induce tissue alterations in kidneys. The study on naturally infected subjects confirmed the presence of renal alterations in a high percentage of the examined animals (18/21). Eleven cases presented glomerular changes (three glomerulonephritis and 8 glomerular amyloidosis) and tubulointerstitial lesions, and were therefore considered severe alterations. The other seven subjects presented milder glomerular changes that consisted in mesangial-matrix expansion with or without segmental glomerulosclerosis.

Previous studies with naturally FIV-infected cats suggested significantly higher rates of renal dysfunction and histological changes compared to FIV seronegative ones. Examinations of 326 sick cats from Australia demonstrated a significant association between FIV infection and azotemia and palpably small kidneys [27]. Concerning small kidneys, as noted by abdominal palpation, they were also reported by Brown and colleagues [28]. Nonspecific renal abnormalities were also reported by several studies [13,29]. The percentage of FIV infected cats with similar renal alteration ranged from 5.5% in New Zealand [14] to 9.3% (from a survey of 700 cats) in Japan [15]. In 76 cats from three regions in Italy (Piemonte, Liguria and Val d'Aosta) 9% were affected by renal disease [30]. Previous histopathological and ultrastructural investigations described kidney abnormalities in 12 of 15 cats [16] and 10 of 14 cats [17] with naturally acquired FIV infection. Six of the twelve subjects of the first study presented lesions that caused a marked increase in serum BUN and creatinine concentration and heavy glomerular non-selective proteinuria; the other nine cats with renal abnormalities, the urine protein content was higher than normal range (>0,2g/l) [16]. Results obtained in the present study showed similar findings. 18 of 21 (82%) of naturally infected cats had mild to sever proteinuria (mean value of 26.46±22.41 (ranging from 3.5 – 62 g/L). 10/21 cats had glomerulo non-selective proteinuria, 8/21 of them combined with tubular proteinuria.

386 Immunodeficiency

**3.4. Immunohistochemistry** 

Immunohistochemistry Control

with or without segmental glomerulosclerosis.

IgG deposits in

uninfected controls.

**4. Discussion** 

IgG deposits in capilalry

The results of immunohistochemical investigations are summarized in table 5. By IF and IHC positive specimens showed segmental, predominantly granular mesangial deposits of IgG, IgM and C3, while rarely scattered granular deposits were detected along the capillary loops (Figure 1C). IgA staining was never observed. Cellular infiltrates were characterized by the presence of IgG secreting plasma cells and scattered IgM. Large proteinaceous casts were positive for IgG and weakly for IgA. Amyloid deposits were always positive for the mouse monoclonal anti-human AA and the rabbit polyclonal against the feline AA amyloid

> Naturally FIV-infected cats

mesangium 0/4 3/21 3/17 5/28 3/11

loops 0/4 1/21 1/17 2/28 2/11 IgM deposits 0/4 14/21 6/17 6/28 3/11 IgA deposits 0/4 0/21 0/17 0/28 0/11 C3 deposits 0/4 14/21 6/17 6/28 3/11 **Table 5.** Main immunohistochemical findings in naturally and experimentally FIV-infected cats and

The results of our investigation clearly demonstrate that FIV infection induce tissue alterations in kidneys. The study on naturally infected subjects confirmed the presence of renal alterations in a high percentage of the examined animals (18/21). Eleven cases presented glomerular changes (three glomerulonephritis and 8 glomerular amyloidosis) and tubulointerstitial lesions, and were therefore considered severe alterations. The other seven subjects presented milder glomerular changes that consisted in mesangial-matrix expansion

Previous studies with naturally FIV-infected cats suggested significantly higher rates of renal dysfunction and histological changes compared to FIV seronegative ones. Examinations of 326 sick cats from Australia demonstrated a significant association between FIV infection and azotemia and palpably small kidneys [27]. Concerning small kidneys, as noted by abdominal palpation, they were also reported by Brown and colleagues [28]. Nonspecific renal abnormalities were also reported by several studies [13,29]. The percentage of FIV infected cats with similar renal alteration ranged from 5.5% in New Zealand [14] to 9.3% (from a survey of 700 cats) in Japan [15]. In 76 cats from three regions in Italy (Piemonte, Liguria and Val d'Aosta) 9% were affected by renal disease [30]. Previous histopathological and ultrastructural

Experimentally FIV-infected cats Petaluma Pisa M2 Petaluma +

Pisa M2

(Figure 1F), while were always negative for the rabbit polyclonal anti-AL amyloid.

cats

The investigations carried out on experimentally FIV-infected SPF cats demonstrated renal alterations partially similar to those detected in naturally FIV-infected ones. Particularly, mesangial widening with or without segmental glomerulosclerosis and immune mediated glomerulonephritis were observed in these subjects, no matter that they were infected with different FIV isolates, maintained in isolation units, and sacrificed at different times postinfection.

Mesangial widening with segmental to diffuse glomerulosclerosis [16], nephrosclerosis [29] and thickened Bowman's membrane [28] have been previously described in naturally FIVinfected cats. These alterations represent glomerular reactions common to many apparently unrelated, clinical entities that are currently believed to result from intraglomerular hemodynamic alterations [31]. In FIV-infected cats hemodynamic alterations might be mediated by sustained production of lymphokines and/or factors of mesangial proliferation with activity on glomerular capillary permeability as a consequence of the chronic systemic viral infection. Although controversial, increasing evidence supports a direct effect of the virus on renal cells either as a result of exposure to viral proteins or direct renal parenchyma infection. The use of a HIV-1 transgenic mouse model demonstrated a direct etiologic link between HIV- 1 expression in kidney and the development of segmental glomerulosclerosis in HIV associated nephropathy (HIVAN) with unique viral-host interactions, which depends on inherent features of the virus and, at the same time, host response [32]. In FIV infection the direct role of the virus in the pathogenesis of renal alterations is postulated by the presence of p24 viral antigen in tubular epithelial cells as well as scattered interstitial inflammatory and glomerular cells and by detection of FIV *gag* DNA and RNA sequences in these subjects [17,33].

Even if FIV-infected cats often present hypergammaglobulinemia, which is believed to be triggered by chronic polyclonal B cell activation [34], which, in turn, can lead to the production of immune complexes [34,35] and auto-antibodies [36], immune complex glomerulonephritis are infrequent. In a previous study on 15 naturally FIV-infected cats only one subjects showed IgG deposits in mesangial areas [16). In this study IgG deposits were detected in 3/21 naturally and in 11/56 experimentally FIV-infected cats, associated with segmental and focal mesangioproliferative glomerulonephritis in 13 cases and only with a membranoproliferative glomerulonephritis. Even if, as mentioned, immune mediated glomerulonephritis seem uncommon in FIV-infected cats, previous studies demonstrated that the mean concentration of circulating IgG immune complex (CIC) in FIV-infected cats were significantly higher than in control cats, while IgM levels increased only slightly. The immunoglobulin fractions from 10/15 renal tissue samples were analysed, found to be polyclonal, and only partly specific for FIV antigens. All these results, including hypergammaglobulinemia and high levels of CIC, together suggest that IC might play a role in the pathogenesis of the renal alteration observed in FIV-infected animals [34]. The evidence of focal and granular deposits of IgM and C3 in mesangium and sclerotic loops of the other subjects is likely the result from nonspecific trapping of serum proteins rather than from immune complex deposition [16].

Feline Immunodeficiency Virus (FIV) Infection in Cats: A Possible Cause of Renal Pathological Changes 389

alteration were aproteinuric. (>2.0 g/L). Significantly higher mean values were established in 12 cats with glomerular alteration; mean value of uP in this group was 21.53±29.39 g/L (2.8 – 100.0) and UPC ration was 4.8±5.77 (0.90 – 17.63). All 20 cats with tubular alteration had glomerulonephritis (9/20) or mensangial widening (11/20). Three of them were aproteinuric. Only four of six cats, with interstitial infiltrates were proteinuric, two 24 months post-infection and 4 after over 30 months post inoculation. In addition, electrophoresis of urine proteins confirmed the correlation between proteins excreted in the urine and the histological alterations found in observed cats; three of 21 cats had glomerulo selective proteinura, 15

The overwhelming majority of experimentally infected cats (71.4 to 100%) had inverted CD4+/CD8+ T cell ratio that depended on infecting viral isolate and, albeit with low or no statistical significance, time of infection. Furthermore, no correlation between CD4+/CD8+ T cell ratio and renal alteration was found. As well as clinical and laboratory findings, as azotemia and proteinuria, renal disease in HIV positive patients seem to be similar in FIV

There are many reports about various organ system involvements, including renal in HIVinfected people [41]. Patient with HIVAN may develop a spectrum of renal pathology that most likely manifests with an acute rapidly progressive loss of renal function, characterised by proteinuria, nephrotic syndrome and azotemia [27]. Typical histological features consist of focal sclerosing glomerulopathy and microcystic tubular dilatation. Some of the patients have mesengial proliferation [27,41,42]. The pathogenesis of HIVAN is still unclear, although presence of viral proteins in glomerular and tubular epithelium cells suggested an important role of HIV in the initiation or progression of HIVAN [27]. Glomerular manifestations include

In conclusion, our study confirmed that renal involvement occurs in a high proportion of naturally FIV-infected cats and that these alterations can, in part, be detected in experimentally infected subjects. In all, these results suggest a causative relationship between FIV infection and renal abnormalities. This damage seems to consist of mesangial increase, sometimes accompanied by mesangial cell proliferation and glomerulosclerosis, a lower percentage of immune mediate glomerulonephritis and, in naturally infected subjects' glomerular and interstitial amyloidosis. Besides, these alteration are partially similar to those detected in HIV-infected patients, FIV-infected cats might represent an interesting natural animal model for the study of the pathogenesis of HIV-associated nephropathy and

*Clinic for Small Animal Medicine and Surgery, Veterinary Faculty, University of Ljubljana,* 

infected, which do not correlate with CD4+ T-cell count or CD4+/CD8+ T cell ratio [28].

antigen-antibody complex and nonimmune-complex-mediated pathology [43,44].

renal alterations associated with chronic viral infection.

**5. Conclusion** 

**Author details** 

*Ljubljana, Slovenia* 

Natasa Tozon

glomerulo non-selective, three glomerulo non-selective and tubular proteinuria.

Tubulo-interstitial lesions consisting of interstitial infiltration by lymphocytes and plasma cells, as well as fibrosis and tubular degenerative changes have been detected in a high proportion of naturally FIV-infected cats [16,28,29,37], but rarely in experimentally infected ones (17/21 vs 6/56, respectively).

Our study confirmed that glomerular and interstitial amyloidosis can be observed in the kidney of naturally FIV-infected animals as previously reported [17,37,38], but amyloid deposits were no detected in the renal tissues of the 56 experimentally infected cats examined. Histochemical and immunohistochemical studies demonstrated that amyloid deposits were consistent with secondary amyloidosis, associated with chronic infections. Previous unpublished data, demonstrated that naturally FIV-infected cats had a higher prevalence of renal amyloidosis compared to uninfected subjects, 12/34 naturally FIVinfected cats vs 1/30 age-matched control cats. The data of this study show that FIV infection alone is not sufficient for the development of amyloid deposition, as demonstrated by the absence of amyloidosis the 56 experimentally FIV-infected cats and that other concurrent factors are needed.

In cats naturally and experimentally infected with FIV have been reported the presence of renal lymphoid tumours [39,40]. In our series we have no cases of lymphosarcoma, but the lack of these neoplastic alterations could be related to the reduced number of subjects examined.

Clinicopathological studies revealed the relatively high possibility of mild to sever renal proteinuria without clinical signs of azotemia. None of 56 SPF cats included in our study presented any clinical signs of azotemia except one cat that was infected by Pisa-M2, sacrificed one year post-infection and that showed serum creatinine concentration (144 μmol/L) with mild proteinuria (3.9 g/L and UPC 0.43) (CRF stage 2) [26]. Renal proteinuria, established by UPC was present in 10 of 17 cats infected with Petaluma, 9 of 28 cats, infected with Pisa-M2 and 2 of 11 in cats infected with both virus isolates, regardless the time from infection. However, the most severe renal proteinuria was observed in cats infected with Pisa-M2 (mean 3.27±6.34; ranged 0.53-17.63), slightly milder in cats infected with both virus strains (mean 3.0±4.39; ranged 0.55-13.01) and the less severe in cats infected with Petaluma (mean 2.56±3.03; ranged 0.31 – 7.02). Fourteen of 29 cats without any histological renal alteration observed had the lowest mean value of urine protein concentration (uP) 3.95±1.37 g/L (2.5 – 7.05) and the lowest UPC mean value of 0.45±0.12 (0.25-0.67). The mean values of uP and UPC in 10 proteinuric cats with mesangial widening was slightly higher 3.54±1.52 g/L (2.0 – 5.4) and 0.81±0.44 (0.53 – 1.94), respectively. Five cats with similar mesangial alteration were aproteinuric. (>2.0 g/L). Significantly higher mean values were established in 12 cats with glomerular alteration; mean value of uP in this group was 21.53±29.39 g/L (2.8 – 100.0) and UPC ration was 4.8±5.77 (0.90 – 17.63). All 20 cats with tubular alteration had glomerulonephritis (9/20) or mensangial widening (11/20). Three of them were aproteinuric. Only four of six cats, with interstitial infiltrates were proteinuric, two 24 months post-infection and 4 after over 30 months post inoculation. In addition, electrophoresis of urine proteins confirmed the correlation between proteins excreted in the urine and the histological alterations found in observed cats; three of 21 cats had glomerulo selective proteinura, 15 glomerulo non-selective, three glomerulo non-selective and tubular proteinuria.

The overwhelming majority of experimentally infected cats (71.4 to 100%) had inverted CD4+/CD8+ T cell ratio that depended on infecting viral isolate and, albeit with low or no statistical significance, time of infection. Furthermore, no correlation between CD4+/CD8+ T cell ratio and renal alteration was found. As well as clinical and laboratory findings, as azotemia and proteinuria, renal disease in HIV positive patients seem to be similar in FIV infected, which do not correlate with CD4+ T-cell count or CD4+/CD8+ T cell ratio [28].

There are many reports about various organ system involvements, including renal in HIVinfected people [41]. Patient with HIVAN may develop a spectrum of renal pathology that most likely manifests with an acute rapidly progressive loss of renal function, characterised by proteinuria, nephrotic syndrome and azotemia [27]. Typical histological features consist of focal sclerosing glomerulopathy and microcystic tubular dilatation. Some of the patients have mesengial proliferation [27,41,42]. The pathogenesis of HIVAN is still unclear, although presence of viral proteins in glomerular and tubular epithelium cells suggested an important role of HIV in the initiation or progression of HIVAN [27]. Glomerular manifestations include antigen-antibody complex and nonimmune-complex-mediated pathology [43,44].

### **5. Conclusion**

388 Immunodeficiency

from immune complex deposition [16].

ones (17/21 vs 6/56, respectively).

factors are needed.

examined.

immunoglobulin fractions from 10/15 renal tissue samples were analysed, found to be polyclonal, and only partly specific for FIV antigens. All these results, including hypergammaglobulinemia and high levels of CIC, together suggest that IC might play a role in the pathogenesis of the renal alteration observed in FIV-infected animals [34]. The evidence of focal and granular deposits of IgM and C3 in mesangium and sclerotic loops of the other subjects is likely the result from nonspecific trapping of serum proteins rather than

Tubulo-interstitial lesions consisting of interstitial infiltration by lymphocytes and plasma cells, as well as fibrosis and tubular degenerative changes have been detected in a high proportion of naturally FIV-infected cats [16,28,29,37], but rarely in experimentally infected

Our study confirmed that glomerular and interstitial amyloidosis can be observed in the kidney of naturally FIV-infected animals as previously reported [17,37,38], but amyloid deposits were no detected in the renal tissues of the 56 experimentally infected cats examined. Histochemical and immunohistochemical studies demonstrated that amyloid deposits were consistent with secondary amyloidosis, associated with chronic infections. Previous unpublished data, demonstrated that naturally FIV-infected cats had a higher prevalence of renal amyloidosis compared to uninfected subjects, 12/34 naturally FIVinfected cats vs 1/30 age-matched control cats. The data of this study show that FIV infection alone is not sufficient for the development of amyloid deposition, as demonstrated by the absence of amyloidosis the 56 experimentally FIV-infected cats and that other concurrent

In cats naturally and experimentally infected with FIV have been reported the presence of renal lymphoid tumours [39,40]. In our series we have no cases of lymphosarcoma, but the lack of these neoplastic alterations could be related to the reduced number of subjects

Clinicopathological studies revealed the relatively high possibility of mild to sever renal proteinuria without clinical signs of azotemia. None of 56 SPF cats included in our study presented any clinical signs of azotemia except one cat that was infected by Pisa-M2, sacrificed one year post-infection and that showed serum creatinine concentration (144 μmol/L) with mild proteinuria (3.9 g/L and UPC 0.43) (CRF stage 2) [26]. Renal proteinuria, established by UPC was present in 10 of 17 cats infected with Petaluma, 9 of 28 cats, infected with Pisa-M2 and 2 of 11 in cats infected with both virus isolates, regardless the time from infection. However, the most severe renal proteinuria was observed in cats infected with Pisa-M2 (mean 3.27±6.34; ranged 0.53-17.63), slightly milder in cats infected with both virus strains (mean 3.0±4.39; ranged 0.55-13.01) and the less severe in cats infected with Petaluma (mean 2.56±3.03; ranged 0.31 – 7.02). Fourteen of 29 cats without any histological renal alteration observed had the lowest mean value of urine protein concentration (uP) 3.95±1.37 g/L (2.5 – 7.05) and the lowest UPC mean value of 0.45±0.12 (0.25-0.67). The mean values of uP and UPC in 10 proteinuric cats with mesangial widening was slightly higher 3.54±1.52 g/L (2.0 – 5.4) and 0.81±0.44 (0.53 – 1.94), respectively. Five cats with similar mesangial In conclusion, our study confirmed that renal involvement occurs in a high proportion of naturally FIV-infected cats and that these alterations can, in part, be detected in experimentally infected subjects. In all, these results suggest a causative relationship between FIV infection and renal abnormalities. This damage seems to consist of mesangial increase, sometimes accompanied by mesangial cell proliferation and glomerulosclerosis, a lower percentage of immune mediate glomerulonephritis and, in naturally infected subjects' glomerular and interstitial amyloidosis. Besides, these alteration are partially similar to those detected in HIV-infected patients, FIV-infected cats might represent an interesting natural animal model for the study of the pathogenesis of HIV-associated nephropathy and renal alterations associated with chronic viral infection.

#### **Author details**

Natasa Tozon *Clinic for Small Animal Medicine and Surgery, Veterinary Faculty, University of Ljubljana, Ljubljana, Slovenia* 

Mauro Pistello *Department of Experimental Pathology, Medical Biotechnologies, Infectious Diseases and Epidemiology, University of Pisa, Pisa, Italy* 

Alessandro Poli *Department of Animal Pathology, Prophylaxis and Food Hygiene, Veterinary Faculty, University of Pisa, Pisa, Italy* 
