Schistosomiasis

Chronic parasitic worm infection with Schistosoma haematobium primarily affects the genitourinary tract (i.e., urogenital schistosomiasis) and can result in hematuria (bloody urine), urinary symptoms, genital sores increasing HIV risk, bladder and ureteral fibrosis (scarring), and bladder cancer.  Consequently, S. haematobium may be the most lethal worm on the planet. Our research group is particularly interested in combining microsurgical and in vivo imaging techniques with experimental models of urogenital schistosomiasis to better understand the immunology and epithelial and stromal biology of this terrible disease. S. haematobium is endemic in Africa and the Middle East, and relies on freshwater snails as an obligatory part of its life cycle. Controlling infestation has proven challenging due to the need to preserve local aquatic ecosystems and difficulties with public education. Currently, the only available medications to treat schistosomiasis are praziquantel and artemether. There is no approved vaccine.

Due to technical issues with animal models, the majority of basic schistosomiasis research is performed on S. mansoni and S. japonicum, the major causes of the intestinal form of schistosomiasis. However, over 112 million people are infected with S. haematobium, the most common schistosome species globallyDiagnosis of S. haematobium infection (urogenital schistosomiasis) is made by manual microscopic counting of eggs in urine. This method is expensive, labor-intensive, and impractical in many endemic areas, given the lack of consistent electricity needed to power modern microscopes. Urogenital schistosomiasis-induced urinary tract fibrosis and cancer can require surgery, even after clearance of infection through medication. Coupled with the possibility of emerging schistosomal resistance to praziquantel, there is a need for vaccines, alternative diagnostic tests and medications, and a better understanding of the immunopathology associated with S. haematobium.

The life cycle of human schistosomes:

schistosomiasis_life_cycle_700

Source: Wikipedia entry for schistosomiasis

To view a 2012 Stanford Bio-X video of Professor Hsieh discussing schistosomiasis and the lab’s work click here.

Lab video footage of an S. haematobium cercariae:

Lab video footage of S. haematobium worms:

Lab video footage of an S. haematobium miracidium:

Our overarching hypothesis on urogenital schistosomiasis is that Schistosoma haematobium eggs induce tissue-specific immune responses that result in compromised host defense and urothelial de-differentiation. These responses render hosts susceptible to co-infections such as HIV and urinary tract infections, as well as bladder carcinogenesis.

Next >

Innovative Approaches to Biological Barriers

Urogenital schistosomiasis, infection by parasitic Schistosoma haematobium worms, persists as a global scourge in part because of a lack of experimentally manipulable animal models of this disease. Mice are cheap, easy to handle, and feature many species-specific tools compared to other animal models, but their small size can prove challenging for certain experimental tasks. Moreover, natural transdermal infection of mice with S. haematobium does not consistently result in urogenital disease. Using microsurgical techniques developed in our laboratory, we have successfully collaborated with Dr. De’Broski Herbert at the University of Cincinnati to develop more reliable mouse models of S. haematobium infection:

banner-micro-surgeryBladder wall microinjection of mice with S. haematobium eggs results in chronic inflammation very similar to that seen in human urogenital schistosomiasis. The techniques involved are described in detail in our JoVE article (see video below) and in Fu et al. Our hope is that these novel mouse models will accelerate new diagnostic and therapeutic modalities for urogenital schistosomiasis, as well as enhance fundamental knowledge of this global disease. For example, we are working with Drs. Conor Caffrey and Joseph Liao to develop better diagnostics for this infection.

Note: If you are having trouble viewing the video below please update your Flash Player. If this is not the problem, please visit the article at the JoVE website.

S. haematobium research has also lagged behind work on other parasites due partly to difficulties with establishing stably transgenic schistosomes. We are optimistic that recent advances in comparative genomics, functional genomics, and tractable animal models will facilitate explorations of the interactions of S. haematobium with host systems:

j-infect-dis-trifecta

For instance, data from high-throughput genomic, proteomic, and transcriptomic sequencing can be annotated through functional genomic tools. Transgenesis would facilitate forward genetics by insertional mutagenesis, that, in turn, can be analyzed by genomic approaches. Genetically modified parasites would be tested in tractable rodent models, and parasites obtained from these models would be studied using genomic approaches and functional tools. Hypotheses derived from comparative genomics would be tested using functional genomic tools and animal models of urogenital schistosomiasis (From Rinaldi et al.)

So how will be generate stably transgenic schistosomes? The figure below shows one possible successful strategy:

j-infect-dis-stable-transgenesis

Cartoon depicting an approach to derive and maintain stable lines of transgenic schistosomes. A female schistosome releasing in vitro laid eggs exposed to murine leukemia virus retroviral virions ( particles in upper left hand panel) and cultured with or without antibiotic. Miracidia hatched from eggs would be used to infect Bulinus truncatus snails. Cercariae cultured (through so-called in tandem antibiotic selection) with or without antibiotic could be screened for the presence of the transgene, for transgene copy number, and for expression. The transgenic line(s) could be propagated and maintained in hamsters. When the transgenic line emerges (eg, a line expressing an small hairpin RNA targeting a gene of interest), the eggs would be collected and analyzed in vitro by cellular and molecular assays and in vivo assays, including mouse models of urogenital schistosomiasis (adapted from Rinaldi et al). Image under “In vivo assays” shows bladder wall injection with Schistosoma haematobium eggs (as described by Hsieh et al). The egg bolus can be seen as a semiopaque bleb localized to the bladder wall (arrow in bottom panel). From Rinaldi et al.

If murine leukemia viral-based approaches do not work, there are additional alternatives to establishing transgenic schistosomes:

 

j-infect-dis-venn

Methods to facilitate transgenesis for Schistosoma haematobium. General schema to establish transgenesis for schistosomes, in which(1) culture conditions of developmental stages, (2) genomic DNA transformation strategies (ie, retroviral transduction), and (3) stableexpression of transgenes are linked to (4) specific selection conditions to specifically enrich the population of transgenic worms. Adapted from Chamberlin, published in Rinaldi et al.

< Prev    Next >

Watching Biology in Action

The advent of high resolution imaging modalities such as micro-ultrasonography has revolutionized biomedical research. In our efforts to experimentally recapitulate urogenital schistosomiasis (infection by parasitic Schistosoma haematobium worms), we have employed micro-ultrasonography to serially monitor S. haematobium egg-induced mouse bladder granuloma development in vivo:

Video footage of micro-ultrasound probe scrolling along the z-axis of the lower abdomen of an egg-injected mouse demonstrates the presence of a bright, echogenic (dense) round granuloma impinging on the upper right side of the urine-filled (black), otherwise ovoid bladder lumen. From Fu et al.

We are also exploring other in vivo imaging modalities such as 2-photon microscopy:

 

< Prev    Next >

figure-1_dgVolcano plots of differentially transcribed genes in egg-injected bladders. Separate plots are shown comparing microarray-derived gene transcription in egg-injected bladders relative to control vehicle-injected bladders (A, one week post-injection; B, three weeks post-injection; C, five weeks post-injection). Log fold changes are plotted on the x-axes and negative log10 p-values are plotted on the y-axes. Red and other non-dark gray, colored symbols denote statistically significant (p<0.05) changes in gene transcription. Gray symbols denote statistically insignificant (p>0.05) changes in change transcription. Legend for individual, selected genes of interest applies to all three panels. For clarity, each selected gene of interest with multiple microarray probes has been denoted using the probe featuring the greatest differential signal for that gene. Adapted from Ray et al., PLOS Neglected Tropical Diseases

We have also used the Genes2WordCloud web site to generate a word cloud based on correlations of our differentially transcribed genes to Generif annotations:

generif-genes2wordcloud-for-2012-plos-ntd-paper

Expected terms included the tissue-associated words “collagen”, “epithelial”, and “muscle”. Other expected terms encompassed those related to carcinogenesis, such as “invasion”, “carcinoma”, “mutation”, and “cancer”. Interestingly, a handful of specific immune- and extracelular matrix remodeling-relevant genes are prominent in the word cloud, namely IL6, NFKB, and MMP9. This highlights both broad and narrow themes amongst genes differentially transcribed in the bladder during urogenital schistosomiasis.

Given that the majority of basic schistosomiasis research is performed on Schistosoma mansoni, a major cause of the intestinal form of schistosomiasis, a relevant issue is whether the transcription of immune response- and fibrosis-related genes triggered by S. haematobium eggs in the bladder differ from those directed against S. mansoni eggs in other tissues. Perhaps the most appropriate comparisons can be made between our data and microarray analyses that have employed the S. mansoni egg-induced, synchronous lung granuloma model. These studies are methodologically analogous to this study’s microarray analysis of our synchronous egg injection model. Our data indicates that although the S. haematobium egg-directed immune and fibrotic response in the bladder shares many similarities to the S. mansoni egg-triggered lung response, there are a number of potentially important disparities (see Venn diagram immediately following this paragraph). This highlights the critical need to develop in vivo models which properly match schistosome species with their tropism for specific host organs.

Egg shells are highly autofluorescent, which aids in their identification by a variety of imaging modalities.

Other high resolution techniques we are using to study urogenital schistosomiasis include gene expression microarrays. We have found that S. haematobium eggs induce a complex bladder gene response that waxes and wanes (see figures following this paragraph and Ray et al., PLOS Neglected Tropical Diseases). Few genes are differentially transcribed one week after bladder injection with S. haematobium eggs. However, the number of differentially transcribed genes peaks by three weeks after egg injection of the bladder. By five weeks post-egg injection, the pool of differentially transcribed genes was already contracting. This suggests that the chronic bladder changes seen in urogenital schistosomiasis cannot be sustained by a single set of eggs; rather, it is driven by continuous oviposition by adult worms. In this model, successive waves of egg laying, rather than any lone egg bolus (such as that featured in our model), would sustain a long-term bladder response. This is consistent with observations that early stage schistosomal urinary tract pathology eventually resolves after praziquantel therapy-induced worm death (which leads to cessation of egg laying in the host).

<Prev     Next>

Venn diagram showing a subset of mouse genes transcribed in a differential and shared fashion in the S. mansoni egg-exposed lung and S. haematobium egg-exposed bladder. Based on data from Ray et al. (PLOS Neglected Tropical Diseases)

venn-diagram

In collaboration with Chris Contag, we are developing additional high resolution in vivo imaging approaches (besides microarrays) to our models of urogenital schistosomiasis. These techniques not only improve our ability to characterize biological processes, but also achieve the important economic and humane goals of reducing the number of mice needed for our work.

The “Great Eaters”

Macrophages, Greek for “great eaters”, are leukocytes or white blood cells that phagocytose (engulf) and destroy microbes. These cells are found in abundance in Schistosoma haematobium egg-infected tissues, a feature of urogenital schistosomiasis (infection by parasitic S. haematobium worms). In our PLoS Pathogens paper we have demonstrated the presence of CD68-positive macrophages in our experimental model of urogenital schistosomiasis:

greateater

Three S. haematobium eggs (center of figure) are surrounded by brown-stained, CD68-positive macrophages. Adapted from Fu et al.

This suggests that our model recapitulates some of the key histologic features of human disease, namely the presence of perioval macrophages. Nearly all of the immunopathology associated with urogenital schistosomiasis is believed to derive from the host immune response to eggs deposited in pelvic organs. The classic lesion of schistosomiasis is the granuloma, a collection of “of mature macrophages…accompanied by accessory features such as necrosis or the infiltration of other inflammatory leukocytes” (Adams DO. The granulomatous inflammatory response. Am J Pathol 1976:84:163-192). Thus, macrophages and other granuloma-associated leukocytes are likely to be key players in schistosomal organ pathogenesis. We are characterizing these potential roles for leukocytes in urogenital schistosomiasis through collaborations with Ajay Chawla,De’Broski HerbertAjit Varki, and others.

One approach we have taken to characterize macrophages in our model of urogenital schistosomiasis is microarray-based (Ray et al., PLOS Neglected Tropical Diseases). We have observed increased transcription of general macrophage-specific genes such as MPEG1 and CD68 as well as genes implicated in alternative activation of macrophages (IL-4, arginase-1, CHI3L3, and mannose receptor [MRC1]). Interestingly, our microarray data also reveals increased transcription of the decoy receptor IL13RA2, which can dampen IL-13-dependent alternative activation of macrophages, and IL10RA, a receptor for IL-10 which may foster the development of “suppressive” or M3 macrophages:

Increased transcription of selected macrophage-related genes identified through microarray analysis

macrophage-microarray-gene-graph

*All values ≥2-fold and p<0.05
**Value <2-fold and/or p≥0.05

Adapted from Ray et al., PLOS Neglected Tropical Diseases

 

 

arg1 il4

cd68 mcp-1_ccl2

Real-time PCR corroboration of macrophage gene transcription profiles obtained by microarray analysis. Transcription of a subset of genes analyzed by real-time PCR is depicted in the scatter plots and is displayed as fold change relative to control-injected bladders at 1, 3 and 5 weeks post-injection. Colored bars correspond to microarray data, with decreased, increased, and unchanged transcription colored green, red, and black, respectively. Increasing intensity of green and red bars indicates more extreme changes in transcription (up to >20-fold). * = p<0.05, ** = p<0.01, ***p<0.001 in comparison to control vehicle-injected bladders (or across time points if annotated with brackets). Adapted from Ray et al., PLOS Neglected Tropical Diseases

 

< Prev    Next >

monocyte-and-its-surface-markers

Graphical depiction of our microarray data showing differential transcription of multiple genes involved in monocyte/macrophage function (marked with red stars, data based on Ray et al., PLOS Neglected Tropical Diseases).

We have more directly confirmed a role for macrophages in urogenital schistosomiasis through use of liposomal clodronate, a compound which depletes macrophages:

clodronatechangesgranulomacells_0

Macrophage depletion of egg-injected bladders leads to decreased granuloma infiltration by multiple nonmacrophage leukocyte subsets. Mice were killed on day 14 after egg injection and treatment with clodronate-loaded or control liposomes and bladder harvested, homogenized into single cell suspensions, and subjected to flow cytometry specific for B cells, T cells, macrophages, alternatively activated macrophages (AAM), eosinophils (Eos), and neutrophils (Neu). *p<0.05; **p<0.01; ***p<0.005. From Fu et al.

A critical functional role for macrophages was confirmed through the observation that liposomal clodronate depletion of macrophages in S. haematobium egg-injected bladders results in egg dose-dependent mortality over time, which is particularly striking given that mice were only exposed to a single bolus of eggs:

survivalcurve

Mice underwent intraperitoneal injection with liposomes and bladder wall injection with liposomes mixed with different doses of S. haematobium eggs. *p<0.05; **p<0.01; ***p<0.005. From Fu et al.

When we examined the bladders of these mice, we noted that macrophage-depleted, egg-injected bladders featured severe hemorrhage, suggesting that macrophages play a key part in preserving bladder integrity during urogenital schistosomiasis:

bladderhemorrhage

Representative photographs are shown of bladder explants taken 2 days after bladder wall injection of mice with eggs or control vehicle and control (empty) or clodronate-loaded liposomes. From Fu et al.

We suspect that macrophages not only preserve bladder integrity during urogenital schistosomiasis, but may also prevent translocation of bacterial endotoxin from the urine in the bladder into the bloodstream, given that S. haematobium egg-injected mice feature higher systemic endotoxin levels when they have been depleted of macrophages using liposomal clodronate:

endotoxin

Endotoxemia develops in macrophage-depleted, egg-injected mice. Sera were collected 2 days postegg or vehicle injection from mice that underwent clodronate or control liposome treatment. The sera were subjected to limulus amebocyte lysate (LAL) assays to measure endotoxin levels. *p<0.05. Horizontal bars indicate mean values. From Fu et al.

We have also uncovered an important role for macrophages in female genital schistosomiasis (FGS), the form of disease caused by S. haematobium eggs depositing in the female reproductive organs. FGS purportedly increases HIV risk through unknown mechanisms. One theorized mechanism is that FGS results in macrophage and T cell infiltration of the vaginal submucosa, the very cells that are targets for HIV infection through expression of the CXCR4 and CCR5 coreceptors for the virus. Through mouse vaginal wall injection with S. haematobium eggs we have been able to generate submucosal lesions that resemble FGS:

vaginalgranuloma2wksx4

Adapted from Richardson et a., PLOS Neglected Tropical Diseases, 2014.

 

< Prev    Next >

When we made single cell suspensions from the vaginal tissues of S. haematobium egg-injected mice, we noted extensive numbers of macrophages and CXCR4+ and CCR5+ T cells:

vaginalflow

Adapted from Richardson et a., PLOS Neglected Tropical Diseases, 2014.

Hence, it is plausible that FGS contributes to HIV infection risk by increasing the numbers of potential HIV target cells in the tissues of the female reproductive tract.

Studying the Microbial-Host Interface

Urogenital schistosomiasis, infection by parasitic Schistosoma haematobium worms, induces bladder urothelial (inner cellular lining) changes, including bladder cancer. One of the precursor lesions of bladder cancer is urothelial hyperplasia (cellular proliferation). This is a prime example of microbial-host interactions, and we have demonstrated that our mouse model of urogenital schistosomiasis leads to urothelial hyperplasia:

8-d99-sh-eggs-40xurotheliumBladder wall microinjection of mice with S. haematobium eggs induces early and sustained urothelial hyperplasia with reactive nuclear changes. From Fu et al.

We are actively developing models of urogenital schistosomiasis-associated bladder cancer in collaboration with Mark Gonzalgo. Other areas of inquiry related to microbial-host interactions at the urothelial interface include mechanisms of hematuria (bloody urine) and alterations in urothelial barrier and immune functions. Some of these projects are joint ventures with Philip BeachyAlan Pao, and Mark Nicolls.

With regards to urothelial barrier functions, our microarray data (Ray et al., PLOS Neglected Tropical Diseases) indicates that a single exposure to S. haematobium eggs triggers significant changes in transcription of urothelial function-related genes. This includes decreased transcription of all uroplakin genes, in addition to several tight junction-related genes (see figures immediately following this paragraph). Collectively, these urine barrier function genes are critically important for the ability of the bladder to safely store and expel urine. We speculate that S. haematobium may have evolved to compromise the human host bladder barrier in order to facilitate egg expulsion into the urine and thus propagation of the life cycle. Indeed, we have observed shed eggs in the urine of experimentally infected animals:

aileens-eggs-102011-day13

S. haematobium eggs shed in the urine of an egg-injected mouse. Adapted from Ray et al., PLOS Neglected Tropical Diseases.

uroplakin-microarray-graph

Microarray analysis reveals urothelial barrier function genes with less transcription after egg exposure. Graph shows fold regulation post-injection relative to control. All data shown p<0.05. Based on Ray et al., PLOS Neglected Tropical Diseases.

 

< Prev    Next >

We have also corroborated these microarray findings through real-time PCR (Ray et al., PLOS Neglected Tropical Diseases, differential transcription of genes is displayed below as fold change relative to control-injected bladders. Colored bars correspond to microarray data, with decreased, increased, and unchanged transcription colored green, red, and black, respectively. * = p<0.05, ** = p<0.01, ***p<0.001):

upk1a upk1b upk2 upk3a upk3b cldn8

These findings, namely epithelial hyperplasia in the setting of compromised barrier function, are consistent with strategies used by a number of helminths to manipulate hosts for their own gain:

wormholes_400

Adapted from Boyett and Hsieh, PLOS Pathogens, 2014

S. haematobium infection also compromises the bladder urothelium by inducing cancer through poorly understood mechanisms presumably related to chronic inflammation. Our model may facilitate a better understanding of the inflammatory pathways leading to schistosomal bladder carcinogenesis, and perhaps cancers in general. For example, gene expression microarray analysis (Ray et al., PLOS Neglected Tropical Diseases) of the S. haematobium egg-exposed mouse bladder demonstrated differential transcription of multiple genes implicated in cancer, including oncogene-, mammary carcinogenesis(breast cancer induction)-, and vascular endothelial growth-factor-related pathways:

vegf-gene-pathway

Pathways analysis reveals differential transcription of multiple gene members of the vascular endothelial growth factor-related pathway five weeks post-egg exposure. All genes marked with red stars featured ≥2-fold differential transcription and p<0.05.

We have also discovered that a single exposure of the mouse bladder to S. haematobium eggs triggers massive changes in DNA methylation of the urothelial genome:

methylome

Genome map demonstrating that multiple loci are hypermethylated (magenta dots) and hypomethylated (green dots) in the mouseurothelium exposed to S. haematobium eggs versus vehicle controls. Representative map from one of two experimental replicates shown. From Conti et al.

When we performed pathways analysis on differential methylation events present in the egg- versus vehicle-exposed bladder urothelium, we found that the former featured differential methylation events in multiple members of the Wnt signaling pathway, including the bladder cancer-associated tumor suppressor gene Wif1:

wnt

S. haematobium eggs induce differential methylation of multiple members of the Wnt signaling pathway, including the bladder cancer-associated tumor suppressor gene Wif1. Differentially methylated gene members of the Wnt signaling pathway are circled in red. Note thatWif1, a tumor suppressor gene implicated in bladder carcinogenesis, is differentially methylated and sits far upstream along the Wnt pathway.Figure generated using DAVID  (http://david.abcc.ncifcrf.gov/). From Conti et al.

 

< Prev    Next >

Although this data was suggestive of a causal role for DNA methylation in the precancerous changes that may lead to schistosomal bladder cancer, it is well known that DNA methylation events can be “passengers”, i.e., incidental changes, rather than “drivers” of biological phenomena. To establish a causal relationship between the observed DNA methylation events and preneoplastic alterations of the S. haematobium-exposed bladder (i.e., urothelial hyperplasia, a necessary but not sufficient event in carcinogenesis), we administered the DNA methylation inhibitor 5-fluoro-2′-deoxycytidine (FdCyd) and tetrahydrouridine (THU), an inhibitor of FdCyd metaboslim, to S. haematobium egg-injected mice:

fdcyd

In vivo inhibition of DNA methylation prevents S. haematobium egg-induced urothelial hyperplasia, a potential preneoplastic lesion of thebladder. (A) Every other day administration of FdCyd (12.5 mg/kg) and THU (25 mg/kg) for 14 days prevents S. haematobium egg-inducedurothelial hyperplasia (n = 4 mice, right panel) compared to vehicle control (n = 3 mice, left panel). The yellow and black arrows denote thethickness of the urothelium overlying the egg-induced bladder granuloma. (B) Bar graph depicting urothelial thickness from (A) in graphicalformat (“DNMTI” and “Vehicle” indicate the DNA methylation inhibitor- and vehicle-treated s, respectively). From Conti et al.

Understanding Internal Scarring

11-d35-10x-tcUrogenital schistosomiasis, infection with parasitic Schistosoma haematobium worms, kills 150,000 people annually through urinary tract fibrosis (internal scarring)-induced kidney failure. This makes S. haematobium perhaps the deadliest worm globally. We have confirmed that our mouse model of urogenital schistosomiasis recapitulates the profound bladder fibrosis found in many people with chronic infection:

S. haematobium egg-injected mouse bladders demonstrate histologically-apparent fibrosis within granulomata (Masson’s trichrome stain, blue staining shows abnormal deposition of collagen, a hallmark of fibrosis). From Fu et al. 

Inflammation leading to tissue fibrosis is the end result of many diseases besides urogenital schistosomiasis, including renal failure, autoimmune diseases such as systemic sclerosis, and certain chronic infections. These pathological processes support the concept of the immune system as a key regulator of tissue remodeling. We are investigating the means by which this may occur in urogenital schistosomiasis. Some of our work on these stromal effects of urogenital schistosomiasis are collaborations with Philip Beachy. This work may identify common immune mechanisms of fibrosis that may serve as therapeutic targets.

The complexity of mechanisms of fibrosis in our model of urogenital schistosomiasis is hinted at by our microarray data (Ray et al., PLOS Neglected Tropical Diseases). Numerous genes linked to extracellular matric remodeling and fibrosis are differentially transcribed in the S. haematobium egg-exposed mouse bladder:

ecm-microarray-graph

 

Microarray analysis reveals extracellular matrix-related genes with differential transcription after egg exposure. Graph shows fold regulation post-injection relative to control. All data shown p<0.05. Based on Ray et al., PLOS Neglected Tropical Diseases.

We have also corroborated these microarray findings through real-time PCR (Ray et al., PLOS Neglected Tropical Diseases, differential transcription of genes is displayed below as fold change relative to control-injected bladders. Colored bars correspond to microarray data, with decreased, increased, and unchanged transcription colored green, red, and black, respectively. * = p<0.05, ** = p<0.01, ***p<0.001):

col17a1 col3a1 col4a5

We have also linked macrophage activity to bladder fibrosis in our mouse model of urogenital schistosomiasis. When S. haematobium egg-injected mice are depleted of macrophages using liposomal clodronate, their bladders feature significantly disrupted collagen deposition and altered granuloma architecture:

massons

Masson’s Trichrome staining is shown of mouse bladders injected 21 days earlier with S. haematobium eggs and treated with either control liposomes or liposomal clodronate. From Fu et al.

 

<Prev