Lymphangiogenesis Is Upregulated in Kidneys of Patients With Multiple Myeloma.код для вставкиСкачать
THE ANATOMICAL RECORD 293:1497–1505 (2010) Lymphangiogenesis Is Upregulated in Kidneys of Patients With Multiple Myeloma JULIA K. ZIMMER,1 SUZAN DAHDAL,2 CHRISTIAN MÜHLFELD,3 IVO P. BERGMANN,1 MATHIAS GUGGER,2 AND UYEN HUYNH-DO1* 1 Department of Nephrology and Hypertension, Inselspital, University of Bern, Bern, Switzerland 2 Institute of Pathology, University of Bern, Bern, Switzerland 3 Institute of Anatomy, University of Bern, Bern, Switzerland ABSTRACT Neolymphangiogenesis has recently been demonstrated in transplanted kidneys as well as in chronic interstitial nephritis and IgA nephropathy. However, its signiﬁcance in kidney disease remains to be deﬁned and a systematic study of renal lymphangiogenesis is warranted. We investigated patients with multiple myeloma (MM) presenting in the great majority with acute renal insufﬁciency. Controls were allograft kidney donors and patients with renal insufﬁciency due to acute renal failure (ARF). Lymph vessel length density (LVD) was quantiﬁed immunohistochemically by means of antipodoplanin staining followed by computer-assisted stereology. The mean LVD in kidneys of patients with MM (23.19 mm2) was higher when compared with allograft donors (7.42 mm2, P ¼ 0.0003) and patients with ARF (6.78 mm2, P ¼ 0.0002). The higher LVD was signiﬁcantly associated with interstitial inﬂammation, and the newly formed lymph vessels were accompanied by diffuse and nodular interstitial inﬁltrates composed mainly of CD20þ B cells and CD27þ plasma cells. The inﬁltrates in patients with MM also displayed a higher expression of the B-cell chemoattractant CXCL13. These results demonstrate for the ﬁrst time that lymphangiogenesis is a prominent feature in MM kidneys and that it is associated with a signiﬁcant accumulation of macrophages, CD20þ and CD27þ B lymphocytes. Further studies should clarify whether these changes represent a beneﬁcial or detrimental factor in the progression of the myeloma-related kidney damC 2010 Wiley-Liss, Inc. age. Anat Rec, 293:1497–1505, 2010. V Key words: lymphangiogenesis; multiple myeloma; cast nephropathy; B cells; chemokines Multiple myeloma (MM) belongs to the group of plasma cell dyscrasias and around four new cases per 100,000 members of the general population are diagnosed every year (Kyle et al., 2004), with a peak at the age of 68 years. It is one of the main reasons for unexplained renal failure in elderly people (Haas et al., 2000). About 50% of patients with MM show increased serum creatinine at diagnosis and 15%–20% already have signiﬁcantly impaired renal function (serum creatinine 177–221 lmol/L or 2.0–2.5 mg/dL). The most frequent histological changes in kidney biopsies of patients with MM are cast nephropathy (40%–63%), light chain deposition disease (LCDD, 19%–26%), and amyloidosis C 2010 WILEY-LISS, INC. V These data were presented as a poster at the 40th Annual Meeting of the ASN in San Francisco. Grant sponsor: Swiss National Science Foundation; Grant number: 3100A0-105871. *Correspondence to: Uyen Huynh-Do, MD, MME, Department of Nephrology and Hypertension, Inselspital, University of Bern, CH-3010 Bern, Switzerland. Fax: þ41-31-632-9734. E-mail: email@example.com Received 15 October 2009; Accepted 18 March 2010 DOI 10.1002/ar.21189 Published online 20 July 2010 in Wiley Online Library (wileyonlinelibrary.com). 1498 ZIMMER ET AL. (7%–30%; Montseny et al., 1998). Renal affection and remaining function have an important impact on the prognosis of the patient. Patients with cast nephropathy have an especially poor survival outcome once renal insufﬁciency has progressed to end-stage renal disease requiring dialysis. Therefore, it is crucial to identify MM-related renal diseases as early as possible to initiate adequate treatment. Potential therapeutic strategies for improving renal function are chemotherapy, with or without subsequent autologous stem cell transplantation, and extended dialysis with protein-leaking dialyzers (Hutchison et al., 2007). According to in vitro and in vivo data (Arimura et al., 2006), the application of the pituitary adenylate cyclase-activating polypeptide could be a promising alternative in the future. Angiogenesis has become a major topic in tumor research due to its role in metastatic spread and its potential prognostic value. List (2001) reported an upregulation of vascular endothelial growth factor (VEGF) in MM, while others have found an increased microvessel density in bone marrow of patients with MM induced by VEGF or hepatocyte growth factor (Alexandrakis et al., 2004). These observations are the rationale behind targeting MM with antiangiogenic drugs such as thalidomide (Du et al., 2004). In contrast, lymphangiogenesis has only very recently become the topic of investigation following the discovery of reliable and speciﬁc lymphatic endothelial markers (Sleeman et al., 2001). The lymphatic endothelial cells of these thin-walled vessels can be detected with the following antibodies: anti-Prox-1, anti-LYVE-1, and antipodoplanin (Breiteneder-Geleff et al., 1999). Prox-1 is a key transcription factor for the lymphatic phenotype, and LYVE-1 is a hyaluronate receptor that ﬁrst appears during lymphatic development, while the mucoprotein podoplanin is probably responsible for the shape of podocytes (Matsui et al., 1999). Besides well-established functions such as ﬂuid balance and immune cell transport, little is known about the role of lymphatic vessels in the kidney and other parenchymal organs. Lymph vessels originate from embryonic veins, and except for the eye and the brain, they can be found physiologically in several organs as a characteristic feature of higher vertebrates. This ‘‘second circulation’’ network is also present in the kidney, where the lymph vessels are conﬁned to the area around largeand middle-sized arteries (McIntosh and Morris, 1971) in the cortex, but do not appear in the peritubular space (Kerjaschki et al., 2004). The medulla of a healthy person rarely possesses any lymphatics (Cuttino et al., 1989). Just a few years ago, the Kerjaschki group showed that neolymphangiogenesis occurs in allograft kidney transplants, a ﬁnding conﬁrmed by another report 3 years later (Adair et al., 2007; Stuht et al., 2007). In another study, Kerjaschki et al. (2006) demonstrated that lymphatic endothelial cells in renal transplants originated from recipient-derived lymphatic progenitor cells. These studies on lymphangiogenesis in renal transplants were followed by a recent report demonstrating neolymphangiogenesis in chronic interstitial nephritis as well as in IgA nephropathy (Heller et al., 2007). Those lymph vessels were associated with nodular lymphatic inﬁltrates. To date however, and to the best of our knowledge, none of these studies have used morphometric methods to quantify renal neolymphangiogenesis. De novo formation of tertiary lymphoid organs (‘‘lymphoid neogenesis’’) with typical B lymphocyte accumulation is also present in multiple autoimmune diseases such as rheumatoid arthritis (Takemura et al., 2001) or inﬂammatory bowel disease (Kaiserling, 2001). Martin and Chan (2006) pointed out that B cells are involved in tissue injury via chemokine release and Tcell response triggering. MM can induce several non-MM speciﬁc as well as speciﬁc lesions (i.e., amyloidosis, cast nephropathy, LCDD) in the kidney depending on the stage of the disease. Cast nephropathy or amyloidosis is mainly displayed in the full-blown picture of late myeloma phases. The aim of this retrospective study was to use rigorous morphometric methods to determine whether an upregulation of lymphangiogenesis is present in MM kidneys and if there is a difference between the early and late disease stages. A further hypothesis was that kidneys of patients with MM are associated with immunological alterations such as tissue inﬁltrations of immunocompetent cells. By characterizing the composition of these inﬁltrates, we looked for potential triggers of renal lymphangiogenesis. MATERIALS AND METHODS Study Population Archival renal biopsies from patients with MM who presented in the great majority with an acute decline of kidney function were used (n ¼ 37). The diagnosis of typical myeloma signs in the kidney, that is, cast nephropathy, LCDD, or amyloidosis, was based on light microscopy, immunohistochemistry (IHC), immunoﬂuorescence, and/or electron microscopy. Fifteen biopsies from healthy kidney allograft donors taken prior to transplantation and 12 biopsies from patients with acute renal failure (ARF) due to other causes served as controls. Data collection and usage were performed with informed patient consent and in accordance with the Declaration of Helsinki. Immunohistochemistry and Immunoﬂuorescence IHC and immunoﬂuorescence were performed on 3lm-thick formalin-ﬁxed and parafﬁn-embedded tissue samples. Slides were dewaxed and hydrated in xylole and ethanol before being microwaved in 10 mM Tris– 1 mM EDTA buffer, pH ¼ 9.0 (410 W, 2 min 5 min). Lymphatic vessels were identiﬁed by staining with monoclonal mouse anti-human podoplanin IgG (D2-40, 1:1,000; Covance, Lausen, Switzerland). Since the ﬁrst description of its use in 1997 (Breiteneder-Geleff et al., 1997), this marker has been shown in numerous studies to allow a speciﬁc discrimination between lymphatic and vascular vessels in a very reliable manner. Furthermore, it works very well for IHC on parafﬁn sections (Fogt et al., 2004; Galambos and Nodit, 2005; Kerjaschki et al., 2006). Characterization of inﬁltrating cell populations and chemokines was performed using the following antibodies: monoclonal mouse anti-human CD68 (PGM1, 1:1,000) and CD20 (L26, 1:1,000; Dako, Baar, Switzerland), CD27 (137B4, 1:50; Dianova, Hamburg, Germany), CD3 (F2.2.38, 1:100; Antikoerper-online, Aachen, Germany), and goat anti-human CXCL13 (polyclonal, RENAL LYMPHANGIOGENESIS IN MULTIPLE MYELOMA 1:100; R&D, Abingdon, UK). For IHC, a biotin-streptavidin-alkaline phosphatase-based method was applied according to manufacturer’s instructions (LSABþ kit, New Fuchsin, Dako). Immunoﬂuorescence labeling was done with mouse anti-human podoplanin (monoclonal, 1:50, Dako) and monoclonal mouse anti-Ki-67 (SP6, 1:100, Thermo RM-9106-S1). Appropriate secondary antibodies labeled with ﬂuorochromes Alexa 488 and Alexa 546 (Molecular Probes, Luzern, Switzerland) were used, and the stainings were visualized with a Zeiss Axiophot 2 microscope. Analysis of Lymph Vessel Length Density Length density of lymph vessels was quantiﬁed at an objective lens magniﬁcation of 20 using a light microscope equipped with a camera and a computer-assisted stereology tool (CAST, Olympus Denmark). Systematic uniform random sampling was applied to ensure that every part of the section had an equal chance of being included in the analysis. An unbiased counting frame with an area of 21668.7 lm2 was projected onto each test ﬁeld. Lymph vessels were recognized by positive immunostaining against podoplanin, and their proﬁles were counted if they were lying inside the counting frame, but not touching the exclusion line or its extensions. Length density of lymph vessels was calculated from the number of lymph vessel proﬁles, the number of counting frames, and the area of the counting frame according to the following formula (Weibel, 1979): Lv ðlymph vessels=biopsyÞ 2 Q Possible cornersCF ¼ ACF CornersCF on tissue where Q is the lymph vessel proﬁles; possible cornersCF ¼ 4; CF is the counting frame; and A is the area of CF. Length density refers to the length of a structure contained within a unit of the reference volume and is therefore expressed in mm2. As length density estimations may change, either due to the length of a structure or to variations in the reference volume, they should ideally be multiplied by the reference volume, which gives rise to the total length. However, it was not possible to gather data on the reference volume (i.e., total kidney volume) of the patients. To ensure that there had been no signiﬁcant changes in the reference volume, the kidney ultrasound data of the different groups were collected. The craniocaudal pole diameter is generally used to describe kidney size. As we found no alteration in this parameter between patients with MM and the control group, we concluded that changes in the length density were due to changes in lymph vessel length and not due to a different reference volume. Therefore, length density provided a suitable parameter for comparing lymph vessel length between the groups. Assessment of Interstitial Fibrosis and Inﬂammation All biopsies were screened for interstitial ﬁbrosis in the Trichrom Masson and for inﬂammation in the hematoxylin and eosin staining at an objective lens magniﬁca- 1499 tion of 10. The evaluation was performed semiquantitatively in a blind manner by two independent investigators (J.Z. and S.D.) according to the score previously used by Heller et al. (2007); for ﬁbrosis, 1: ﬁbrotic tissue <1/3 of the biopsy, 2: ﬁbrotic tissue 1/3< <2/3 of the biopsy, 3: ﬁbrotic tissue >2/3 of the biopsy; for inﬂammation, 1: few cells, diffusely spread, 2: at least one inﬁltrate, 3: conﬂuent inﬁltrates. Analysis of Cellular Inﬁltrates The occurrence of different cell populations and chemokine expression, based on immunostaining with the markers CD20, CD27, CD68, CD3, and CXCL13, was assessed at an objective lens magniﬁcation of 10 in a blind manner by two independent investigators (J.Z. and C.M.). The following semiquantitative score was used: (no positively stained cells), þ (few positively stained cells, diffusely spread), þþ (aggregation of positively stained cells), and þþþ (many aggregations of positively stained cells, partially conﬂuent; Adair et al., 2007). Statistical Analysis Data between groups were analyzed with the KruskalWallis test and the Mann-Whitney U-test as there was no Gaussian distribution. A value of P < 0.05 was considered to be statistically signiﬁcant. RESULTS Increased Lymph Vessel Length Density in Kidneys of Patients With MM Between 1994 and 2007, 37 patients with MM, presenting in the great majority with acute worsening of kidney function, underwent renal biopsy after the treating physicians had excluded reversible causes such as dehydration, hypercalcemia, exposure to nonsteroidal anti-inﬂammatory drugs (NSAIDS) or contrast media, or postrenal obstruction. Their average creatinine was 359 lmol/L, and the average creatinine clearance added up to 30 mL/min. The clinical features of these patients are depicted in Table 1. In 22 patients, histological changes were typical for MM, that is, cast nephropathy (n ¼ 14), LCDD (n ¼ 4), both (n ¼ 1), or amyloidosis (n ¼ 3), whereas only non-MM-speciﬁc alterations were found in 15 biopsies. Fifteen healthy kidney allograft donors (mean creatinine 81 lmol/L) and 12 patients with ARF (mean creatinine 604 lmol/L) served as controls (histological diagnosis in Table 2). By immunostaining for podoplanin expressing LVs, the LV length density was determined in these biopsies, and a higher frequency of small caliber LV in patients with MM versus kidney allograft donor controls was found (Fig. 1). The MM group had a mean LV length density of 23.19 mm2 when compared with 7.42 mm2 in the allograft donor samples (P ¼ 0.0003). To ensure that these ﬁndings were not just due to renal insufﬁciency, a second control group was chosen consisting of patients with ARF due to a variety of etiologies such as acute postinfectious glomerulonephritis, drug-induced interstitial nephritis, or tubular necrosis (creatinine 604 lmol/L vs. 359 lmol/L in MM). With a mean value of 6.78 mm2, LV length density of the ARF group was in the same range as the allograft 1500 ZIMMER ET AL. TABLE 1. Clinical features of patients with multiple myeloma No Gender/Age MM with speciﬁc signs 1 W/56 2 M/80 3 W/61 4 M/77 5 M/63 6 M/73 7 W/75 8 W/62 9 M/80 10 M/61 11 M/74 12 W/58 13 M/68 14 M/60 15 W/78 16 M/72 17 M/61 18 M/72 19 M/58 20 M/47 21 M/43 22 W/54 MM but no speciﬁc signs 23 M/33 24 M/42 25 W/53 26 M/74 27 M/70 28 M/51 29 W/63 30 W/43 31 M/57 32 M/52 33 M/69 34 M/73 35 M/79 36 M/63 37 M/67 Lv (mm2) Histological diagnosis Creatinine (lmol/L) Creatinine clearance (mL/min) Pole distance (ultrasound, in cm) 27.2 53.5 60.9 10.9 12.2 19.5 3.8 19.9 26.5 16.1 22.2 33.4 37.2 37.2 10.2 4.5 7.5 37.6 16.1 11.4 6.0 12.4 CAST Amyloidosis LCDD Amyloidosis CAST CAST LCDD LCDD CAST CAST Amyloidosis CAST CAST CAST LCDDþCAST LCDD CAST CAST CAST CAST CAST CAST 215 – 335 61 976 800 189 142 622 160 478 408 278 234 339 135 854 350 – 1360 390 145 – – 16 – 3 – 29 – 12 42 – – 8 – 44 59 7 22 14 – 11 45 – 11.0 10.2 – – 11.8 – 11.0 10.0 11.0 – 13.0 10.5 12.6 – 12.4 12.0 10.0 13.1 13.8 14.0 11.1 46.6 6.7 44.5 37.9 56.4 24.5 28.9 11.3 7.2 16.6 0.0 24.8 25.9 28.5 11.9 Minor tubular lesions Glomerulosclerosis Minor glomerular changes Membranoprolif.GN Mesangial sclerosis Ischemic damage Glomerulosclerosis Minor arteriosclerosis Tubulointerstitial nephritis Endocapillary proliferation GN Tubulointerstitial nephritis Chronic GN Hypertensive nephropathy Nephangiosclerosis Nephangiosclerosis 450 658 – 200 285 161 183 – 170 164 413 280 176 82 168 – – – – – 66 – 47 50 54 6 3 25 60 – 14.0 12.5 – – – – 10.4 – 12.8 – 10.0 11.6 – – – donors (ARF versus allograft donors, P ¼ 0.30) and signiﬁcantly lower than that of patients with myeloma (P ¼ 0.0002; Fig. 2). Renal ﬁbrosis was assessed semiquantitatively as an additional means of ruling out that higher LV length density in patients with MM could only be due to interstitial ﬁbrosis. We found that the higher lymph vessel length density (LVD) in MM samples was not associated with a higher degree of ﬁbrosis (P ¼ 0.215; Fig. 3A). However, the higher LVD in patients with MM was signiﬁcantly associated with interstitial inﬂammation, for example, lymphocyte inﬁltrates (P ¼ 0.019; Fig. 3B). Because the histopathological spectrum of the MM biopsies reached from typical myeloma signs such as cast nephropathy (Fig. 4; ‘‘ARF with typical MM lesions,’’ n ¼ 23) to changes not speciﬁc for MM (Fig. 4; ‘‘MM with ARF but no speciﬁc lesions,’’ n ¼ 11; ‘‘MM with chronic renal failure,’’ n ¼ 3), a comparison of LV length density was made in these three subpopulations. Interestingly, the different clinical types of the disease showed no difference in LV length density. Patients with MM-speciﬁc affection of the kidney had a mean value of 23.19 mm2 vs. 22.10 mm2 in the group with chronic renal failure and versus 23.47 mm2 among the population with nonMM-speciﬁc alterations (P ¼ 0.872 and P ¼ 1.00; Fig. 4). In each population studied, LV length density showed a certain variation, a ﬁnding also observed in the study by Stuht et al. (2007) and which is due to a widespread distribution of lymph vessels in humans. Nevertheless, the overlap of the LV length densities in the different groups was minor. Proliferating Lymph Vessels in Patients With MM The next step was to examine whether the higher LV length density was associated with ongoing lymphangiogenesis or whether the lymphatic beds were already quiescent at the time of investigation. In kidney transplants, Kerjaschki et al. (2006) demonstrated proliferating lymphatics, but this ﬁnding could not be conﬁrmed in the later study of Adair et al. (2007). We performed immunoﬂuorescence staining for podoplanin and Ki-67. Ki-67 is a protein that is solely expressed in the active phases of mitosis but not in the G0 phase and therefore represents a speciﬁc marker for cell 1501 RENAL LYMPHANGIOGENESIS IN MULTIPLE MYELOMA TABLE 2. Clinical features of control group No Gender/Age Allograft Donors 1 W/38 2 M/14 3 M/23 4 M/40 5 M/42 6 M/44 7 M/49 8 M/21 9 M/55 10 M/30 11 M/62 12 M/53 13 M/70 14 M/48 15 M/68 ARF 1 M/75 2 W/55 3 W/61 4 W/50 5 W/55 6 M/27 7 W/70 8 W/82 9 W/65 10 M/70 11 M/58 12 M/49 Lv (mm2) Histological diagnosis 2.1 16.5 7.6 0.0 5.5 0.0 0.7 20.4 23.3 0.7 0.7 5.4 27.0 0.0 1.4 2.0 17.0 4.6 6.1 7.6 1.7 8.7 10.0 5.0 5.0 4.2 9.5 Healthy Healthy Healthy Healthy Healthy Healthy Healthy Healthy Healthy Healthy Healthy Healthy Healthy Healthy Healthy kidney kidney kidney kidney kidney kidney kidney kidney kidney kidney kidney kidney kidney kidney kidney Postinfectious endocapillary GN Postinfectious endocapillary GN Postinfectious endocapillary GN Postinfectious endocapillary GN Postinfectious tubulointerstitial inﬂammation Postinfectious exsudative GN Drug-induced acute interstitial nephritis Drug-induced acute interstitial nephritis Drug-induced acute interstitial nephritis Acute tublar necrosis Acute tublar necrosis Acute tublar necrosis Creatinine (lmol/L) Creatinine clearance (mL/min) Pole distance (ultrasound, in cm) 109 74 73 105 68 68 78 119 50 82 98 – 94 70 40 – – – – – – – – – – – – – – – – – 12.5 – 12.5 – – 10.2 10.5 9.5 10.8 10.3 – 10 10 164 149 419 260 481 876 1000 998 814 878 602 – 44 47 – 29 – – – – – – – – – 9.8 13.3 11.5 9.8 – – – – 13.6 13.0 – Fig. 1. Lymph vessel staining (!) in patients with multiple myeloma (MM) and control group. Immunohistochemistry was performed with antipodoplanin (D2-40) on formalin-ﬁxed and parafﬁn-embedded tissue of MM (A) and kidney allograft donors (B). New Fuchsin staining and hematoxylin counterstaining. Magniﬁcation, 20. proliferation (Gerdes et al., 1983; Scholzen and Gerdes, 2000). By colocalization of these two markers, active proliferation of LVs in our myeloma samples was shown (Fig. 5), which was comparable in magnitude with the ﬁndings shown by Kerjaschki et al. (2006). Higher Occurrence of B Lymphocytes in MM Kidneys As previously described by Kerjaschki et al. (2004) in transplanted kidneys, we found nodular cellular inﬁltrates within the regions of higher LV count. To further characterize the inﬁltrating cell populations, IHC stain- Fig. 2. Increased lymph vessel length density in patients with MM when compared with allograft donors and patients with acute renal failure (ARF). Quantitative analysis of lymph vessel length is shown as scatter plot. The horizontal bar represents the mean. P ¼ 0.0003 between MM and allograft donors; P ¼ 0.0002 between MM and ARF. No signiﬁcant difference between the allograft donor and the ARF group (P ¼ 0.30). ings for B lymphocytes, that is, CD20 for B cells and CD27 for plasma cells, were performed. Figure 6(A,C) clearly demonstrates aggregations of CD20þ and CD27þ B lymphocytes in patients with MM, whereas the control samples showed weaker or even no signals (Fig. 6B,D). 1502 ZIMMER ET AL. Fig. 4. Comparable lymph vessel length density in different subgroups of patients with MM: ARF with typical myeloma-related lesions, ARF with non-MM-speciﬁc lesions and chronic renal failure. Quantitative analysis of lymph vessel length is shown as scatter plot. The horizontal bar represents the mean. There was no signiﬁcant difference between the three groups (P ¼ 1.00, P ¼ 0.872, and P ¼ 0.876). Fig. 3. MM samples with increased LVD do not display increased renal ﬁbrosis but a higher grade of inﬂammatory inﬁltrates. Evaluation of interstitial ﬁbrosis and inﬂammation was performed by two investigators in a blind manner using a semiquantitative score. The horizontal bar in the box plot represents the mean. While the ﬁbrosis level was the same (P ¼ 0.215, SD for MM 0.56 and for control 0.43), inﬂammatory cells were more signiﬁcantly present in patients with myeloma when compared with the control group (P ¼ 0.019, SD for MM 0.50 and for control 0.35). Higher Occurrence of Macrophages in MM Kidneys Macrophages are able to trigger lymphangiogenesis via two pathways: either by release of VEGF-C or by metamorphosis into lymphatic endothelial cells (Maruyama et al., 2005). The parafﬁn slides were immunolabeled with anti-CD68 to determine the number of inﬁltrating macrophages. CD68þ cells were more abundant in patients with MM when compared with controls (Fig. 7, A vs. B). Higher Level of CXCL 13 in MM Kidneys Another mediator that macrophages are capable of producing is the chemokine CXCL13, which has been shown to be a strong chemoattractant for B lymphocytes (Legler et al., 1998). Immunohistochemical staining for CXCL13 conﬁrmed the hypothesis of a higher CXCL13 chemokine level in MM kidneys when compared with controls (Fig. 7, C vs. D). Slight Increase in CD31 Lymphocytes in MM Kidneys To complete the different subsets of cell populations, a semiquantitative analysis of T lymphocytes by means of Fig. 5. Proliferating lymph vessels: colocalization of podoplanin and Ki-67. Immunoﬂuorescence staining for podoplanin (red) and Ki-67 (green) in a patient with MM. The subpart (A) shows a longitudinal section while (B) represents a cross-sectional staining. Magniﬁcation, 40. anti-CD3 was performed. The number of CD3þ cells only slightly increased in MM samples versus controls (data not shown). An overview of all analyzed cell populations and the chemokine CXCL13 is given in Table 3 (patients with MM, n ¼ 34; allograft donors, n ¼ 15). DISCUSSION Research on renal lymphangiogenesis under pathological conditions started only very recently following the discovery of speciﬁc lymphatic markers. In this study, lymph vessels in kidneys of patients with MM were detected by immunohistochemical staining for podoplanin. A pronounced increase in LVD in this patient group was found, accompanied by a prominent tissue inﬁltration by B lymphocytes and macrophages. The demonstration that this higher LVD was not associated with a 1503 RENAL LYMPHANGIOGENESIS IN MULTIPLE MYELOMA TABLE 3. Semiquantitative analysis of cell populations and chemokine expression in patients with MM(n 5 34) and controls (allograft donors, n 5 15) Marker MM Allograft Donors CD20 CD27 CXCL13 CD68 CD3 þþ/þþþ þþ/þþþ /þ þþ/þþþ /þ /þ /þ /þ /þ /þ Evaluation was performed by two investigators in a blind manner using a semiquantitative score (, þ, þþ, þþþ). Fig. 6. Immunostaining for CD20 (A) and CD27 (C) in patients with MM versus controls (B, D). New Fuchsin staining and hematoxylin counterstaining. CD20þ resp. CD27þ lymphocytes are stained as red dots (arrows !). Magniﬁcation, 20. Fig. 7. Immunostaining for CD68 (A, !) and CXCL 13 (C) in patients with MM versus controls (B, D). New Fuchsin staining and hematoxylin counterstaining. CD68þ macrophages are stained as red dots (arrow !). Magniﬁcation, 20. more advanced degree of interstitial ﬁbrosis suggests that these ﬁndings are not just a ‘‘paraphenomenon’’ secondary to interstitial scarrings but that they could be speciﬁcally caused by the disease itself and the associated inﬂammation processes. Most data about renal lymphangiogenesis have so far been obtained in kidney transplant recipients with either acute graft rejection or chronic allograft nephropathy (Kerjaschki et al., 2004; Adair et al., 2007; Stuht et al., 2007). In both cases, the number of lymph vessels was much higher when compared with native nephrectomy tissue; however, there was no signiﬁcant difference in LV density between different Banff rejection grades and chronic allograft nephropathy (Stuht et al., 2007). The signiﬁcance and clinical relevance of lymph vessel presence is still not completely elucidated. It is considered as a detrimental factor in cancer because metastatic dissemination and early lymph node metastasis are associated with a poor prognosis (Morice et al., 2003; Schoppmann et al., 2004). In contrast, the relevance of neolymphangiogenesis in diseases such as asthma or inﬂammatory arthritis (Zhang et al., 2007) remains unclear. This uncertainty is reﬂected in the ongoing post-transplant discussion on whether increased LVD in the kidney allograft predicts a higher risk of organ loss or a better graft survival. Both an ongoing maintenance of immune response through lymphocyte invasion and a clearing function of inﬁltrates can be attributed to lymph vessels (Kerjaschki et al., 2004). Thus, neolymphangiogenesis could be detrimental if the immune process is perpetuated and augmented in the inﬂammed tissues, but it may be beneﬁcial if cytotoxic cells are removed more efﬁciently. In kidney grafts, the newly formed lymphatics reach deep into the tubulointerstitial space, unlike in normal kidneys (Cuttino et al., 1989), and are associated with nodular inﬁltrates that can be found in 10%–15% of transplant biopsies. As apoptosis can only partially remove inﬁltrates at a rate similar to the one in the thymus, lymph vessels could be responsible for the residual evacuation (Paavonen et al., 2000). Their function as so-called exit routes could explain an even better kidney graft function if lymphatics are present within nodular inﬁltrates. This hypothesis is supported by the work from Stuht et al. (2007), who showed that kidney graft function 1 year after transplantation was better in cases with increased lymphangiogenesis. More recently, Adair et al. (2007) also demonstrated new interstitial lymph vessels in nephrectomized end-stage allografts. The authors mentioned that no correlation between lymph vessel number and time to graft failure was found. However, the fact that their study of transplant nephrectomy specimens had its limitations as they represented the end stage of a disease process that was multifactorial in origin was acknowledged. However, what about the presence and role of neolymphangiogenesis in the native kidney? To date, there has been only one study addressing this question. As B cells have recently emerged as potential players and therapeutic targets in inﬂammatory diseases of the kidney, Heller et al. (2007) evaluated biopsies from patients 1504 ZIMMER ET AL. with IgA nephropathy and chronic interstitial nephritis. They found an increased number of CD20þ lymphocytes in these kidneys, together with a striking clustering of neolymphatics around the aggregates of inﬂammatory cells. Control group with acute interstitial nephritis also displayed an increased inﬁltration by CD20þ cells, but neolymphangiogenesis was absent. Obviously, an acute inﬂammatory process alone is not sufﬁcient to trigger renal lymphangiogenesis. Our results conﬁrm this because lymph vessel length of patients with ARF due to postinfectious glomerulonephritis or acute interstitial nephritis did not differ from the allograft donor group. In our patient collective, there was no direct correlation between the number of lymphatics and renal function at presentation. We recognize that our population was heterogenous; however, this has been and will be the problem of all clinicopathological studies dealing with kidney diseases in patients with myeloma. As our knowledge about this subject improves, we realize that ‘‘myeloma kidney" is no longer the sole manifestation of renal involvement in plasma cell dyscrasias, although the common denominator of these seemingly heterogenous clinicopathological pictures remains the detrimental effect of deposited free light chains and the ensuing inﬂammation. The MM collective was further divided into three subgroups for analysis: one subgroup with ARF associated with typical myeloma-related changes in the kidney, a second subgroup with ARF but no myeloma speciﬁc changes, and a third category with MM and chronic renal failure presenting with non-MM-speciﬁc histological alterations. Interestingly, we observed comparable LVD in all three groups. This raises the question of whether the observed changes could be speciﬁc for myeloma. However, two arguments speak against this hypothesis. First, according to our internal guidelines, renal biopsies were performed only when an elevated creatinine persisted after all reversible causes for MM-associated renal insufﬁciency had been ruled out or treated (i.e., treatment of dehydration and/or hypercalcemia, withdrawal from NSAIDS, supportive therapy for acute tubular necrosis (ATN), etc.). Second, as shown in Table 1, the group of patients with MM with non-MM-speciﬁc renal histologies encompassed a wide range of histopathologic diagnoses ranging from minor lesions (n ¼ 3) to predominantly glomerular (n ¼ 6), tubular damage (n ¼ 3), or nephroangiosclerosis (n ¼ 3)—the only common denominator being active MM. In aggregate, this observation suggests that neolymphangiogenesis could be an early feature of MMassociated renal disease. The colocalization experiments with podoplanin and Ki-67 present evidence that this process is due to ongoing lymph vessel proliferation. The factors potentially inﬂuencing neolymphangiogenesis in the kidney were then assessed. Besides the soluble growth factors VEGF-C and VEGF-D (Mandriota et al., 2001), ﬁbroblast growth factor (Kubo et al., 2002), or platelet-derived growth factor (Cao et al., 2004) neolymphangiogenesis can also be inﬂuenced by several cell populations such as B lymphocytes. Angeli et al. (2006) depicted a close interaction between antigen-presenting dendritic cells, B cells, and lymph vessels. Together with T cells, these three components are the main characteristics of the so-called tertiary lymphoid organs, whose occurrence has been described in chronic inﬂammatory processes such as rheumatoid arthritis (Manzo et al., 2005) or in infectious diseases caused by Helicobacter pylori or hepatitis C virus (Freni et al., 1995). Renal lymphoid neogenesis has been observed in kidney allografts underlying chronic rejection (Thaunat et al., 2005). In native kidneys, B lymphocytes have been found more frequently in a number of glomerulopathies, such as IgA or membranous nephropathy (Heller et al., 2007). Our analysis of the cellular inﬁltrates accompanying newly formed lymphatics revealed a clear dominance of B lymphocytes and plasma cells in patients with myeloma when compared with controls. The accumulation of these B lymphocytes was possibly triggered by CXCL13, a chemokine produced by parenchymal cells and macrophages and which has been shown to be a strong B-cell chemoattractant (Legler et al., 1998). Although CXCL13 was nearly absent in control biopsies, an abundance was observed in lymphatic inﬁltrates of MM biopsies. As a higher occurrence of macrophages in the myeloma samples was also detected, we hypothesize a dual role for these cells in renal neolymphangiogenesis: as a consequence of myeloma-related tissue injury, macrophages invade the renal interstitium and release CXCL13. Thereby, B lymphocytes are attracted and, after activation, produce VEGF-A (Angeli et al., 2006), which in turn can also recruit new macrophages. On the other hand, macrophages can also directly stimulate lymphangiogenesis through the release of VEGF-C or metamorphose themselves into lymphatic endothelial cells (Kerjaschki, 2005). In summary, this study shows for the ﬁrst time that lymphangiogenesis is a striking feature in kidneys of patients with myeloma. It is accompanied by a local accumulation of likely triggers such as CD20þ, CD27þ B lymphocytes, and macrophages as can be encountered in chronic processes like interstitial nephritis. Interestingly, LVD was signiﬁcantly higher in MM samples even before disease-speciﬁc kidney damage appeared. This observation, if conﬁrmed by further studies encompassing a larger number of patients, would suggest that lymph vessel staining could be a useful tool for early identiﬁcation of myeloma-related kidney injury. As a consequence, further prospective investigations, including serial biopsies before and after initiation of chemotherapy, will be needed to clarify whether renal neolymphangiogenesis represents a beneﬁcial or a detrimental factor in the progression of the myeloma-related kidney injury. ACKNOWLEDGMENTS The authors thank B.M. Frey and F.J. Frey (Department of Nephrology and Hypertension, Inselspital, University of Bern, Switzerland) for their support and fruitful input. They also thank S. Moll (Renal Pathology, University of Geneva, Switzerland) for some biopsies of patients with multiple myeloma. LITERATURE CITED Adair A, Mitchell DR, Kipari T, Qi F, Bellamy CO, Robertson F, Hughes J, Marson LP. 2007. Peritubular capillary rarefaction and lymphangiogenesis in chronic allograft failure. Transplantation 83:1542–1550. Alexandrakis MG, Passam FJ, Ganotakis E, Dafnis E, Dambaki C, Konsolas J, Kyriakou DS, Stathopoulos E. 2004. Bone marrow microvascular density and angiogenic growth factors in multiple myeloma. Clin Chem Lab Med 42:1122–1126. RENAL LYMPHANGIOGENESIS IN MULTIPLE MYELOMA Angeli V, Ginhoux F, Llodra J, Quemeneur L, Frenette PS, Skobe M, Jessberger R, Merad M, Randolph GJ. 2006. B cell-driven lymphangiogenesis in inﬂamed lymph nodes enhances dendritic cell mobilization. Immunity 24:203–215. Arimura A, Li M, Batuman V. 2006. Potential protective action of pituitary adenylate cyclase-activating polypeptide (PACAP38) on in vitro and in vivo models of myeloma kidney injury. Blood 107:661–668. Breiteneder-Geleff S, Matsui K, Soleiman A, Meraner P, Poczewski H, Kalt R, Schaffner G, Kerjaschki D. 1997. Podoplanin, novel 43kd membrane protein of glomerular epithelial cells, is down-regulated in puromycin nephrosis. Am J Pathol 151:1141–1152. Breiteneder-Geleff S, Soleiman A, Kowalski H, Horvat R, Amann G, Kriehuber E, Diem K, Weninger W, Tschachler E, Alitalo K, Kerjaschki D. 1999. Angiosarcomas express mixed endothelial phenotypes of blood and lymphatic capillaries: podoplanin as a speciﬁc marker for lymphatic endothelium. Am J Pathol 154:385–394. Cao R, Bjorndahl MA, Religa P, Clasper S, Garvin S, Galter D, Meister B, Ikomi F, Tritsaris K, Dissing S, Ohhashi T, Jackson DG, Cao Y. 2004. PDGF-BB induces intratumoral lymphangiogenesis and promotes lymphatic metastasis. Cancer Cell 6:333–345. Cuttino JT, Jr., Clark RL, Jennette JC. 1989. Microradiographic demonstration of human intrarenal microlymphatic pathways. Urol Radiol 11:83–87. Du W, Hattori Y, Hashiguchi A, Kondoh K, Hozumi N, Ikeda Y, Sakamoto M, Hata J, Yamada T. 2004. Tumor angiogenesis in the bone marrow of multiple myeloma patients and its alteration by thalidomide treatment. Pathol Int 54:285–294. Fogt F, Zimmerman RL, Ross HM, Daly T, Gausas RE. 2004. Identiﬁcation of lymphatic vessels in malignant, adenomatous and normal colonic mucosa using the novel immunostain D2-40. Oncol Rep 11:47–50. Freni MA, Artuso D, Gerken G, Spanti C, Maraﬁoti T, Alessi N, Spadaro A, Ajello A, Ferrau O. 1995. Focal lymphocytic aggregates in chronic hepatitis C: occurrence, immunohistochemical characterization, and relation to markers of autoimmunity. Hepatology 22:389–394. Galambos C, Nodit L. 2005. Identiﬁcation of lymphatic endothelium in pediatric vascular tumors and malformations. Pediatr Dev Pathol 8:181–189. Gerdes J, Schwab U, Lemke H, Stein H. 1983. Production of a mouse monoclonal antibody reactive with a human nuclear antigen associated with cell proliferation. Int J Cancer 31:13–20. Haas M, Spargo BH, Wit EJ, Meehan SM. 2000. Etiologies and outcome of acute renal insufﬁciency in older adults: a renal biopsy study of 259 cases. Am J Kidney Dis 35:433–447. Heller F, Lindenmeyer MT, Cohen CD, Brandt U, Draganovici D, Fischereder M, Kretzler M, Anders HJ, Sitter T, Mosberger I, Kerjaschki D, Regele H, Schlondorff D, Segerer S. 2007. The contribution of B cells to renal interstitial inﬂammation. Am J Pathol 170:457–468. Hutchison CA, Cockwell P, Reid S, Chandler K, Mead GP, Harrison J, Hattersley J, Evans ND, Chappell MJ, Cook M, Goehl H, Storr M, Bradwell AR. 2007. Efﬁcient removal of immunoglobulin free light chains by hemodialysis for multiple myeloma: in vitro and in vivo studies. J Am Soc Nephrol 18:886–895. Kaiserling E. 2001. Newly-formed lymph nodes in the submucosa in chronic inﬂammatory bowel disease. Lymphology 34:22–29. Kerjaschki D. 2005. The crucial role of macrophages in lymphangiogenesis. J Clin Invest 115:2316–2319. Kerjaschki D, Huttary N, Raab I, Regele H, Bojarski-Nagy K, Bartel G, Krober SM, Greinix H, Rosenmaier A, Karlhofer F, Wick N, Mazal PR. 2006. Lymphatic endothelial progenitor cells contribute to de novo lymphangiogenesis in human renal transplants. Nat Med 12:230–234. Kerjaschki D, Regele HM, Moosberger I, Nagy-Bojarski K, Watschinger B, Soleiman A, Birner P, Krieger S, Hovorka A, Silberhumer G, Laakkonen P, Petrova T, Langer B, Raab I. 2004. Lymphatic neoangiogenesis in human kidney transplants is associated with immunologically active lymphocytic inﬁltrates. J Am Soc Nephrol 15:603–612. Kubo H, Cao R, Brakenhielm E, Makinen T, Cao Y, Alitalo K. 2002. Blockade of vascular endothelial growth factor receptor-3 signal- 1505 ing inhibits ﬁbroblast growth factor-2-induced lymphangiogenesis in mouse cornea. Proc Natl Acad Sci USA 99:8868–8873. Kyle RA, Therneau TM, Rajkumar SV, Larson DR, Plevak MF, Melton LJ, III. 2004. Incidence of multiple myeloma in Olmsted County, Minnesota: trend over 6 decades. Cancer 101:2667–2674. Legler DF, Loetscher M, Roos RS, Clark-Lewis I, Baggiolini M, Moser B. 1998. B cell-attracting chemokine 1, a human CXC chemokine expressed in lymphoid tissues, selectively attracts B lymphocytes via BLR1/CXCR5. J Exp Med 187:655–660. List AF. 2001. Vascular endothelial growth factor signaling pathway as an emerging target in hematologic malignancies. Oncologist 6:24–31. Mandriota SJ, Jussila L, Jeltsch M, Compagni A, Baetens D, Prevo R, Banerji S, Huarte J, Montesano R, Jackson DG, Orci L, Alitalo K, Christofori G, Pepper MS. 2001. Vascular endothelial growth factor-C-mediated lymphangiogenesis promotes tumour metastasis. EMBO J 20:672–682. Manzo A, Paoletti S, Carulli M, Blades MC, Barone F, Yanni G, Fitzgerald O, Bresnihan B, Caporali R, Montecucco C, Uguccioni M, Pitzalis C. 2005. Systematic microanatomical analysis of CXCL13 and CCL21 in situ production and progressive lymphoid organization in rheumatoid synovitis. Eur J Immunol 35:1347–1359. Martin F, Chan AC. 2006. B cell immunobiology in disease: evolving concepts from the clinic. Annu Rev Immunol 24:467–496. Maruyama K, Ii M, Cursiefen C, Jackson DG, Keino H, Tomita M, Van RN, Takenaka H, D’Amore PA, Stein-Streilein J, Losordo DW, Streilein JW. 2005. Inﬂammation-induced lymphangiogenesis in the cornea arises from CD11b-positive macrophages. J Clin Invest 115:2363–2372. Matsui K, Breitender-Geleff S, Soleiman A, Kowalski H, Kerjaschki D. 1999. Podoplanin, a novel 43-kDa membrane protein, controls the shape of podocytes. Nephrol Dial Transplant 14 (Suppl 1):9–11. McIntosh GH, Morris B. 1971. The lymphatics of the kidney and the formation of renal lymph. J Physiol 214:365–376. Montseny JJ, Kleinknecht D, Meyrier A, Vanhille P, Simon P, Pruna A, Eladari D. 1998. Long-term outcome according to renal histological lesions in 118 patients with monoclonal gammopathies. Nephrol Dial Transplant 13:1438–1445. Morice P, Piovesan P, Rey A, Atallah D, Haie-Meder C, Pautier P, Sideris L, Pomel C, Duvillard P, Castaigne D. 2003. Prognostic value of lymphovascular space invasion determined with hematoxylin-eosin staining in early stage cervical carcinoma: results of a multivariate analysis. Ann Oncol 14:1511–1517. Paavonen K, Puolakkainen P, Jussila L, Jahkola T, Alitalo K. 2000. Vascular endothelial growth factor receptor-3 in lymphangiogenesis in wound healing. Am J Pathol 156:1499–1504. Scholzen T, Gerdes J. 2000. The Ki-67 protein: from the known and the unknown. J Cell Physiol 182:311–322. Schoppmann SF, Bayer G, Aumayr K, Taucher S, Geleff S, Rudas M, Kubista E, Hausmaninger H, Samonigg H, Gnant M, Jakesz R, Horvat R. 2004. Prognostic value of lymphangiogenesis and lymphovascular invasion in invasive breast cancer. Ann Surg 240:306–312. Sleeman JP, Krishnan J, Kirkin V, Baumann P. 2001. Markers for the lymphatic endothelium: in search of the holy grail? Microsc Res Tech 55:61–69. Stuht S, Gwinner W, Franz I, Schwarz A, Jonigk D, Kreipe H, Kerjaschki D, Haller H, Mengel M. 2007. Lymphatic neoangiogenesis in human renal allografts: results from sequential protocol biopsies. Am J Transplant 7:377–384. Takemura S, Braun A, Crowson C, Kurtin PJ, Coﬁeld RH, O’Fallon WM, Goronzy JJ, Weyand CM. 2001. Lymphoid neogenesis in rheumatoid synovitis. J Immunol 167:1072–1080. Thaunat O, Field AC, Dai J, Louedec L, Patey N, Bloch MF, Mandet C, Belair MF, Bruneval P, Meilhac O, Bellon B, Joly E, Michel JB, Nicoletti A. 2005. Lymphoid neogenesis in chronic rejection: evidence for a local humoral alloimmune response. Proc Natl Acad Sci USA 102:14723–14728. Weibel W. 1979. Stereological methods 1. Practical methods for biological morphometry. London: Academic Press. Zhang Q, Lu Y, Proulx S, Guo R, Yao Z, Schwarz EM, Boyce BF, Xing L. 2007. Increased lymphangiogenesis in joints of mice with inﬂammatory arthritis. Arthritis Res Ther 9:R118.