Abstract

Hemosiderosis, the intracellular deposition of iron, is a manifestation of systemic and/or local iron overload. Because naturally occurring hemosiderosis is a common, yet often disregarded, histologic finding in captive Callithricidae, the purpose of this study is to provide a detailed survey of hemosiderosis in a captive callitrichid population. Livers from nine species of callitrichids (Callimico goeldii, Callithrix argentata, Callithrix jaccus, Callithrix pygmea, Saguinus fuscicollis, Saguinus geoffrei, Saguinus midas, Saguinus mystax, Saguinus oedipus) were evaluated for tissue iron concentration, distribution and intensity of intracellular hemosiderin deposition, and degree of fibrosis. The prevalence of hepatic hemosiderosis was 94.4% (n=232), with a high prevalence (97.64%, n=127 neonates examined) of neonatal hepatic hemosiderosis. Mean hepatic iron concentration (HIC) was 4378.2±4459.8 ppm (n=94), ranged from 94.4–18,500 ppm, and correlated positively with the degree of total iron deposition (Spearman rank coefficient of correlation ST, rho=0.811, n=94, p<10^-2 ), hepatocyte iron deposition (ST, rho=0.76, n=94, p<10^-2), and sinusoidal iron deposition (ST, rho=0.73, n=94, p<10^-2). Histologic patterns of hepatic hemosiderosis and dietary analysis strongly suggest that a primary factor influencing hemosiderosis in captive Callithricidae is dietary iron overload. Varying amounts of extrahepatic hemosiderin present in the heart, kidney, lung, adrenal, spleen, pancreas, lymph nodes, intestines, thyroid, testes, and ovary of many animals are likely the result of iron overload. These findings show a high prevalence and severity of hemosiderosis in captive Callithricidae that may be caused by a combination of dietary iron overload, increased bioavailability of iron, and/or a genetic affinity for iron absorption.

Introduction

Hemosiderosis and/or hemochromatosis are common pathologic findings in many captive wild mammals and birds, such as common marmosets²³, captive lemurs¹⁰, Egyptian fruit bats⁶, and mynah birds¹⁵. A local or systemic overload of iron may cause hemosiderosis, the deposition of hemosiderin within cells, and/or hemochromatosis.^5,12,16 Hemochromatosis is tissue iron deposition associated with hepatic hemosiderin deposition, necrosis, fibrosis, nodular regeneration, and damage to other organs, such as the pancreas and heart.^5,12 Spelman et al.²² hypothesize that the clinical disease associated with hemosiderosis in captive lemurs is caused by excessive dietary iron, high dietary ascorbic acid (a factor increasing iron absorption), and low amounts of tannins (a factor decreasing iron absorption). Hemochromatosis associated with excessive dietary iron and ascorbic acid intake is reported as a cause of death in captive Egyptian fruit bats.⁶ Experimental parenteral systemic iron overload in gerbils causes hepatic and cardiac lesions consistent with hemochromatosis.³

While there have been many reports of hemosiderosis in captive callitrichids,^17,18,23 authors differ in their interpretation of its significance and cause. A recent experimental study shows that excessive dietary iron caused common marmosets (Callithrix jacchus) to develop significantly increased liver iron content.¹⁷ In the aforementioned study, marmosets on a high iron diet had a significantly higher mortality rate than control animals. These results suggest that dietary excess of iron may be a significant cause of disease and/or death of captive callitrichids that is commonly overlooked.¹⁷ The purpose of the present study is to evaluate naturally occurring hemosiderosis in captive Callithricidae. The results of this study may be applied to the assessment and improvement of captive management and to the study of diseases in both captive and free-ranging Callithricidae.

Methods

This was a retrospective cross-sectional survey of nine species of captive Callithricidae (Callimico goeldii, Callithrix argentata, Callithrix jaccus, Callithrix pygmea, Saguinus fuscicollis, Saguinus geoffrei, Saguinus midas, Saguinus mystax, Saguinus oedipus) that died at the Bronx Zoo between 1978 and 1997. Samples of formalin fixed livers of 94 callitrichids were sent to Michigan State Diagnostic Laboratory for complete mineral analysis, including liver iron levels. The livers of 232 callitrichids were stained with Perl’s Prussian blue and examined for hemosiderin deposition by two pathologists using a scoring method described by Deugnier et al.⁸ that allowed for the study of the degree of zonal deposition (zones 1, 2, and 3 of Rappaport) of hemosiderin (Prussian blue positive granules) within hepatocytes and mesenchymal tissue (sinusoidal endothelium, macrophages, biliary epithelium, vessels, and connective tissue). According to Rappaport’s definition of the hepatic acinus, zone 1 consists of periportal hepatocytes, zone 2 consists of midzonal hepatocytes, and zone 3 consists of periacinar hepatocytes.¹⁴ Hepatocyte iron score (HIS) is the degree of hemosiderin deposition within hepatocytes. Sinusoidal iron score (SIS) is the amount of hemosiderin within sinusoids. Portal iron score (PIS) is the degree of hemosiderin deposition within portal vessels, connective tissue, and bile duct epithelium. Total iron score (TIS) is calculated as: TIS=HIS+SIS+PIS. Two hundred and nine livers were stained with Masson’s trichrome for detection of hepatic fibrosis. The degree of portal fibrosis was examined by a grading scheme of 0–3, defined as: no portal fibrosis (0), non extensive portal fibrosis (1), extensive portal fibrosis (2), and diffuse hepatic fibrosis with macronodular regeneration (3).⁸ Other histopathologic findings were also noted within livers examined (i.e., triaditis, necrosis, lipidosis). In addition, tissues of selected cases (heart, lung, spleen, adrenal, pancreas, intestinal tract, lymph nodes, thyroid) were examined for hemosiderin deposition. Descriptive and analytic statistical tests were performed using Microsoft Excel for Windows 95 (Version 7.0) and Sigma Stat 2.0 (Jandel Scientific).

Results

Hepatocyte Iron Concentration

Mean HIC was 4378.2±4459.8 ppm (n=94) and ranged from 94.4–18,500 ppm. The mean HIC (ppm) for each species was as follows: C. argentata, 5552.5±3546.8 (n=8); C. goeldii 773.54±714.36 (n=9); C. jacchus, 8423.2±6811.2 (n=15); C. pygmea, 2106.7±1981.8 (n=4); S. fuscicollis, 2388.9±2460.2 (n=12); S. geoffrei, 1710.7±1765.1 (n=7); S. midas, 6036.2±3053.0 (n=8); S. mystax, 5771.9±4833.1 (n=11); S. oedipus, 3893.2±3018.7 (n=17). Among the species sampled, C. goeldii had the lowest mean HIC, 773.54±714.36, and S. midas had the highest mean HIC, 6036.2±3053. HIC differed significantly between Callimico goeldii and the three following species (p<0.05; all pairwise multiple comparison test, Dunn’s method): S. midas, C. jacchus, and C. argentata. There was no significant difference in HIC (ppm) between sex (p=0.084, Mann-Whitney rank sum test) and age class (neonate vs. adult; p=0.33, Mann-Whitney rank sum test).

Histologic Assessment of Hepatic Hemosiderin Deposition

Of callitrichid livers examined, 94.4% (219/232) had some degree of hemosiderin deposition within hepatocytes and/or mesenchymal tissue, ranging from minimal (sparse basophilic dusting) to severe (coalescing masses of hemosiderin). Of callitrichids examined, 5.6% (13/232) had no hemosiderin within liver sections. There was also a high prevalence (97.64%, n=127 neonates examined) of neonatal hepatic hemosiderin deposition. Of callitrichids examined, 17.7% (41/232) had no detectable to mild hepatic hemosiderin deposition (TIS 0–11), and 82.3% (191/232) had moderate to severe hepatic hemosiderin deposition (TIS 12–44). The hepatic iron concentration of animals with no detectable to mild hemosiderin deposition ranged from 95–1,880 ppm and the hepatic iron concentration of animals with moderate to severe hemosiderosis ranged from 94.4–18,500.

Mean TIS was 21.9±10.9 (range: 0–44; n=232). TIS correlated positively with hepatic iron concentration (Spearman rank coefficient of correlation ST, rho=0.811, n=94, p<10^-2). Mean HIS was 16.94±8.86 (range: 0–36; n=232) and mean SIS was 3.8±2.8, n=232). Overall, mean parenchymal iron and sinusoidal hemosiderin deposition was greatest in zone 1, with a decreasing gradient throughout the lobule from zone 1 to zone 3. Both HIS (ST, rho=0.76, n=94, p<10^-2), and SIS (ST, rho=0.73, n=94, p<10^-2), correlated positively with HIC (ppm). There was a slight positive correlation between liver iron concentration and the liver sinusoidal gradient across zones 1 and 3 (ST, rho=0.32, n=94, p<10^-2), with a higher concentration of sinusoidal hemosiderin in zone 1 as liver iron concentration increased. There was no significant correlation between age and HIC, TIS, or HIS. Sinusoidal iron deposition, (ST, rho=0.35, n=209, p<10^-2) correlated positively with age.

There was a significant difference in total hepatocyte hemosiderin deposition and total sinusoidal hemosiderin deposition between adults/subadults (>2 mo of age) and neonates (<2 wk of age) (Mann-Whitney rank sum test, p<0.001). Adults/subadults had higher mean SIS (4.8±3.5) than did neonates (3.1±2.2). Neonates had higher mean total HIS (18.3±7.2) than did adults/subadults (14.7±11.0). There was no significant difference in total liver (hepatocyte and mesenchymal) hemosiderin deposition and overall mesenchymal hemosiderin deposition between adults/subadults and neonates.

Hemosiderin Deposition in Other Tissues and Associated Lesions

There were varying amounts of hemosiderin in the heart, kidney, lung, adrenal, spleen, pancreas, lymph nodes, intestines, thyroid, testes, and ovary. Hemosiderin was present in 19 of 33 (56%) hearts examined. The HIC of animals with intracardiac hemosiderosis ranged from 1,290–18,500 ppm. The HIC of animals with no hemosiderin within their hearts ranged from 434–13,600 ppm. Fourteen of 21 adults/subadults had mild to moderate hemosiderin deposition, primarily within myocytes. Cardiac hemosiderin deposits were found within the sarcoplasm of myocytes, cytoplasm of fibroblasts, and endothelial cells, predominantly in the subepicardial region. Five of 12 neonates examined had mild amounts of cardiac hemosiderin deposition.

Hemosiderin deposition within renal tubular cells and/or renal interstitium was present in 15 of 33 (45%) animals. HIC of animals with renal hemosiderin deposition ranged from 2450–18,500 ppm. Renal hemosiderin deposition was present in 14 of 19 (74%) of adults/subadults examined. Although one neonate out of 14 examined was positive for renal hemosiderin deposition, histologic artifact cannot be ruled out as a cause of renal hemosiderin deposition in this case. Splenic hemosiderin deposition was present within splenic macrophages in 21of 23 (93%) spleens examined. Adrenal cortical hemosiderin deposition was present within cortical cells and connective tissue in 13 out of 19 adrenals examined. Pancreatic hemosiderin deposition was present, primarily within pancreatic interstitium, in five of eleven animals examined. Although no neonates exhibited intestinal hemosiderin deposition, hemosiderin was present within submucosal macrophages of the intestines or colon in 7 of 10 animals. Callitrichids with extrahepatic hemosiderin deposition in three or more organs had a relatively high mean HIC of 8217.7±4717.5 ppm (range: 2,320–18,500 ppm; n=22 callitrichids examined).

Of callitrichids examined, 42.2% (98/232) had hepatic fibrosis ranging from non-extensive portal fibrosis to bridging portal fibrosis. No cases of hepatic cirrhosis were seen within callitrichids examined. Hepatic fibrosis correlated positively with age (ST, rho=0.36, p<0.001, n=209) and total hepatic hemosiderin deposition (ST, rho=0.20, p<0.01, n=232). Although there was no significant correlation between hepatocyte and portal hemosiderin deposition and degree of fibrosis, sinusoidal iron deposition correlated positively with the degree of portal fibrosis (ST, rho=0.36, p<0.001). The degree of portal fibrosis was significantly higher in adults and juveniles than in neonates (Wilcoxon Signed Rank Test, p<0.05). Mild to severe triaditis was found in 37.1% (86/232) of callitrichids. The HIS of callitrichids with triaditis ranged from 95–18,500 ppm. In the sample population, 14.7% (34/232) had multifocal to diffuse hepatic lipidosis.

Discussion

The prevalence, distribution, and severity of hepatic hemosiderosis correlated with hepatic iron concentration in this study strongly suggests that iron overload is a significant pathologic finding in callitrichids that should not be interpreted as innocuous. If large amounts of hepatic iron deposition were normal, then one would expect iron concentrations to be relatively high, with little variation. The large range of hepatic iron concentrations (from 94.4–18,500 ppm) suggests that excessive hepatic iron concentration is abnormal in callitrichids.

The results of this study parallel in many ways studies of human genetic hemochromatosis, as well as experimental animal models of iron overload. Similar to a retrospective study of human genetic hemochromatosis,⁸ this study shows a strong correlation between hepatic iron concentration and total hepatic hemosiderin deposition. In addition, there is a predominance of hepatocellular hemosiderin in both neonatal and adult Callithricidae. Studies show initial parenchymal overload when excessive amounts of iron are absorbed from the gut.¹¹ In humans with hemochromatosis caused by excessive iron absorption (i.e., genetic hemochromatosis), hemosiderin deposits primarily within parenchymal cells of the liver.¹¹ In patients with genetic hemochromatosis, Deugnier et al. (1992)⁸ state that the existence of a decreasing hepatic hemosiderin gradient from zone 1 to zone 3 “strongly points to an intestinal hyperabsorption of iron.” Similarly, in this study, iron tended to accumulate in a zonal gradient throughout the hepatic lobule, with highest average hemosiderin deposition in periportal parenchymal cells (zone 1). Sinusoidal iron deposition in callitrichids correlates positively with age and hepatic iron concentration. In experimental rat models, hepatic sinusoidal iron increases with exposure time to dietary overload.²¹ Thus, the predominant intrahepatocytic hemosiderin deposition, zonal gradient of hepatic hemosiderin, and age-related increase in sinusoidal hemosiderin deposition suggest that hemosiderosis in callitrichids at the Bronx Zoo is primarily due to enteric iron absorption.

In addition, the majority of neonatal callitrichids surveyed had moderate to severe hemosiderin deposition, primarily within hepatic parenchyma. In human neonates, moderate to severe hemosiderosis, similar to intensities seen in this study, is a considered abnormal because most iron stores should be mobilized for hematopoiesis (Dr. Baergen, Cornell Medical College, personal communication). The high prevalence of neonatal hemosiderosis may be due to transfer of excess circulating iron from the maternal plasma to the placental trophoblast, where fetal transferrin binds and carries iron to fetal tissues.⁴

In humans there appears to be a threshold for hepatic iron storage which, when exceeded, results in iron deposition in extrahepatic sites such as the pancreas, heart, kidney, reproductive and endocrine tissue.¹ Hemosiderin is also present in extrahepatic callitrichid tissues, including the heart, adrenal, pancreas, intestines, testes, and ovaries. Callitrichids with extrahepatic hemosiderin deposition in three or more tissues have a high average HIC (8217.7±4717.5 ppm), suggesting that thresholds for hepatic iron storage were exceeded. In humans with systemic iron overload, cardiac failure with associated myocardial hemosiderin deposition is among the most significant causes of death.^2,9 However, the clinical pathology of extrahepatic hemosiderin deposition in callitrichids in this study is not known.

Regarding associated pathology, the degree of periportal and sinusoidal fibrosis was positively correlated with hepatic iron concentration. The extensive hepatic fibrosis often seen in human patients with iron storage disease6,^14,16 and gerbils with parenteral iron overload³ is not present in callitrichids at the Bronx Zoo. The lack of cirrhosis in callitrichids with severe hepatic hemosiderosis should not be interpreted as non-pathologic, because cytosiderosis can cause severe and toxic ultrastructural alterations within hepatocytes.^12,13

Iron overload may also cause immunosuppression by interfering with polymorphonuclear cell superoxide anion generation, phagocytosis, and killing, and by decreased proportions of CD4+ lymphocytes.⁷ Infections with Klebsiella pneumonia, E. coli, Pasteurella multocida, and Clostridium perfringens have been associated with increased available iron.²⁴ Although the present study was not designed to investigate the relationship between hemosiderosis and cause of callitrichid mortality, many deaths in this study sample, as well as in more recent cases, have been attributed to the aforementioned bacterial infections.

Proximate and ultimate factors may account for the high prevalence and intensity of hepatic hemosiderosis in this study. Proximate factors explaining increased hepatic iron storage in callitrichids include dietary iron overload, high dietary vitamin C levels, hypotransferrinemia, and increased intestinal absorption and/or bioavailability of iron. Ascorbic acid in fruits given to the callitrichids may increase bioavailability of iron.²² Estimates of the daily iron concentrations in marmoset and tamarin diets at the Bronx Zoo range from 191.2–238.2 mg/kg and 191.9–305.6 mg/kg, respectively, excluding additional mineral supplementation. Indeed, the diets of captive marmosets and tamarins in this facility contain an iron concentration that exceeds the 180 mg/kg RDA for primates.²⁰ Furthermore, these iron requirements are based on experimental studies in the rhesus monkey, and may be too high for callitrichids.²⁰

Ultimately, callitrichids may have an increased genetic affinity for iron absorption because they have evolved to avidly absorb iron due to living in environments where dietary iron is limited or where supplies of dietary iron are temporally or spatially unpredictable. Interestingly, Callimico goeldii has a significantly lower mean HIC than three species of callitrichids in this study. Perhaps the ecology of Goeldi’s monkeys is such that their foraging strategies include higher dietary iron availability than other callitrichids.

In conclusion, this study’s results show that captive Callithricidae have a high prevalence of hepatic and extrahepatic hemosiderin deposition, supportive of systemic iron overload. Additional captive prospective feeding trial studies, information of mineral intake of free-ranging callitrichids, serum transferrin levels, and ultrastructural studies are necessary to further understand the pathology of hemosiderosis in captive callitrichids.

Acknowledgments

Thanks to A. Ngbokoli, M. Shvarstur, and J. Budde for technical support. We also thank Dr. Baergen, The New York Hospital, Dr. Shilsky, Albert Einstein College of Medicine, Dr. Voelkerding, U. of Wisconsin Hospital, Dr. Benirschke, San Diego Zoological Park, Dr. Trupkiewicz, Philadelphia Zoological Garden, and Dr. Linn, Wildlife Conservation Society, for their professional assistance.

Literature Cited

1.  Adams PC, Y Deugnier, R Moirand, P Brissot. 1997. The relationship between iron overload, clinical symptoms, and age in 410 patients with genetic hemochromatosis. Hepatology. 25(1):162–166.

2.  Barosi G, E Arbustini, A Gavaszzi, M Grasso, A Pucci. 1989. Myocardial iron grading by endomyocardial biopsy. A clinico-pathologic study on iron overloaded patients. Eur. J. Haematol. 42:382–388.

3.  Carthew P, BM Dorman, RE Edwards, JE Francis, AG Smith. 1993. A unique rodent model for both the cardiotoxic and hepatotoxic effects of prolonged iron overload. Laboratory Investigation. 69(2):217–222.

4.  Chase MC, D Riedinger. 1995. Neonatal hemochromatosis: a case report. Neonatal Network. 14(7): 7–12.

5.  Cotran RS, V Kumar, S Robbins. 1994. Robbin’s Pathologic Basis of Disease. 5th ed. Philadelphia: W.B. Saunders.

6.  Crawshaw G, S Oyarzun, E Valdes, K Rose. 1995. Hemochromatosis (iron storage disease) in fruit bats. Pp. 136–146. In: Proceedings of the First Annual Conference of the Nutrition Advisory Group (NAG) of the American Zoo and Aquarium Association (AZA).

7.  De Sousa M. 1989. Immune cell functions in iron overload. Journal of Immunology. 75: 1–6.

8.  Deugnier YM, O Loreal, B Tiurlin, D Guyader, H Jouanolle, R Moirand, C Jacquelinet, P Brissot. 1992. Liver pathology in genetic hemochromatosis: A review of 135 homozygous cases and their bioclinical correlations. Gastroenterology. 102:2050–2059.

9.  Fitchett DH, DJ Coltart, WA Littler, MJ Leyland, T Trueman, D Gozzard, TJ Peters. 1980. Cardiac involvement in secondary haemochromatosis: catheter biopsy study and analysis of myocardium. Cardiovascular Research. 14: 719–724.

10.  Gonzales J, K Benirschke, P Saltman, J Roberts, PT Robinson. 1984. Hemosiderosis in lemurs. Zoo Biology. 3: 255–265.

11.  Iancu, Y Deugnier, JW Halliday, LW Powell, P Brissot. 1997. Ultrastructural sequences during liver iron overload in genetic hemochromatosis. Journal of Hepatology. 27:628–638.

12.  Iancu TC. 1990. Biological and Ultrastructural Aspects of Iron Overload: An Overview.

13.  Iancu TC, HRabinowitz, P Brissot, A Guillouzo, Y Deugnier, M Bourel. 1985. Iron overload of the liver in the baboon: an ultrastructural study. Journal of Hepatology. (1):261–275.

14.  Kelly WR. 1993. The liver and biliary system. Chapter 2. In: KF Jubb, PC Kennedy, N Palmer, eds. The Pathology of Domestic Animals. Volume 2. New York: Academic Press.

15.  Lowenstine & Petrak. 1978. Iron pigment in the livers of birds. Pp. 127–135. In: The Comparative Pathology of Zoo Animals. R Montali, J Migaki, eds. Washington, D.C., Smithsonian Press.

16.  McLaren GD, WA Muir, RW Kellermeyer. 1986. Iron overload disorders: natural history, pathogenesis, diagnosis, and therapy. CRC. Critical Reviews in Clinical Laboratory Sciences. Vol 19(3):204–265.

17.  Miller GF, DE Barnard, RA Woodward, BM Flynn, JWM Bulte. Hepatic hemosiderosis in common marmosets, Callithrix jacchus: effect of diet on incidence and severity. Laboratory Animal Science. 47(2):138–142.

18.  Montali R. 1994. Diseases of zoo marmosets, tamarins, and Goeldi’s monkeys. Proceedings of the ARAV and AAZV. Pittsburgh. Pp.237–239.

19.  Muiesan P, M Rela, P Kane, A Dawan, A Baker, C Ball, AP Mowat, R Williams, ND Heaton.1995. Liver transplantation for neonatal haemochromatosis. Arch. of Disease in Childhood. 73:178v180.

20.  National Research Council. 1978. Nutrient Requirements of Nonhuman Primates. National Academy of Sciences. Washington, D.C.

21.  Park C, B Bacon, GM Brittenham, AS Tavill. 1987. Pathology of dietary carbonyl iron overload in rats. Laboratory Investigation. 57(5): 555–563.

22.  Spelman LH, KG Osborn, MP Anderson. 1989. Pathogenesis of hemosiderosis in lemurs: role of dietary iron, tannin, and ascorbic acid. Zoo Biology. 8:239–251.

23.  Tucker M. 1988. A survey of the pathology of marmosets (Callithrix jacchus) under experiment. Lab. Animals. 18:351–358.

24.  Ward CG, JJ Bullen, HJ Rogers. 1996. Iron and infection: new developments and their implications. Journal of Trauma: Injury, Infection, and Critical Care. 41(2):356–364.