Morphological features of serosa-associated lymphoid clusters of the rat parietal pleura: Exploring a relatively unexplored system.

Cindy G. J. Cleypool and Rosa E. Lagerwerf designed the study, performed data acquisition and analysis, designed the figures, and wrote the manuscript. Claire Mackaaij performed technical procedures and helped with data acquisition and analysis, and with writing the manuscript. Frieke Kuper and Ronald L. A. W. Bleys critically reviewed the manuscript. Cindy G. J. Cleypool supervised the project.

All pleural segments of the parietal pleura, as previously listed in the literature (Table  1 ), were identified in situ (Figure  1 ). In the current study, the DMP is specified as a part of the parietal pleura positioned between the esophagus and aorta (Figure  1 ), whereas the VMP was defined as the part located ventrally to the heart and attaching it to the sternum. The VMP is difficult to distinguish but becomes visible when the sternum and heart are gently pulled apart (Figure  2 ). The existing literature (Table  1 ) lacks a definitive distinction between retropericardial pleural folds (RPF) and the ventral mediastinal pleura (VMP). For clarity, in our study, we designate the pleura located cranially to the adipose streak (a strip of adipose tissue within the RPF, Figure  2 ) as VMP, while the adipose streak and the pleura positioned caudally to it are referred to as RPF. In general, pSALCs can be recognized as small discrete white, cloudy spots on the parietal pleura (Figure  3a illustrates what pSALCs look like in situ without staining). They can occur as solitary structures but mostly occur in clusters and in close association with other structures (nerves, vessels and adipose tissue). After hematoxylin staining, these spots turn bright purple due to their high cellular density of immune cells (Figure  3b illustrates what pSALCs look like in situ after hematoxylin whole mount staining of the thorax). Since RPFs are thin and, except for the adipose streak, contain little to no adipose tissue, their SALCs were most easy to distinguish, even prior to hematoxylin staining. At other locations, such as the part of the RPFs underneath the lung hilum, SALCs were more difficult to discern in situ, even after staining with hematoxylin, where the stained SALCs showed too little contrast with respect to the background. However, after resecting those specific pleural segments from the thorax and placing them in a petri dish, they could be studied with a stereomicroscope with transmitted light, increasing their visibility significantly (Figure  3c illustrates what resected pleural segments look like when studied in a petri dish). All rats showed a high general abundance of pSALCs in their right and left RPF and on their left phrenic nerve, a moderate abundance in their adipose streak and VMP, a low abundance in their left and right pulmonary ligament, and a doubtful presence of pSALCs on their right phrenic nerve. No pSALCs were observed in the left and right pericardial, costal, paravertebral, and diaphragmatic pleura. Table  2 contains an overview on the presence and abundance of pSALCs per pleural segment. During this inspection, it was noticed that based on their morphology, pSALCs can be categorized into four types. The following paragraphs describe the morphological hallmarks of these different types and list their occurrence in the different pleural segments. Stereomicroscopic view of the left retrocardiac pleural fold (RPF) prior and after staining with hematoxylin. (a) Unstained RPF in situ (medial view) with solitary SALCs. SALCs, especially solitary ones, can be recognized as white cloudy spots. The part of the lung that is visible represents the post caval lobe of the right lung. (b) Same RPF as in a after hematoxylin staining. (c) Same RPF as in a and b after resection. SALCs are now more easily recognized without the disturbing interference of other structures. Yellow circles: Similar clusters of solitary SALCs. Presence of pSALCs per pleural segment. Note : All pleural segments were studied for the presence of pSALCs and their relative abundance was evaluated by considering the area of HE stained lymphoid tissue using the following density scale −: complete absence, +: low amount, ++: moderate amount and +++: high amount. In addition, it was listed which types of pSALC were observed. Four different types of pSALCs could be established. Solitary pSALCs occurred as solitary structures in transparent parts of the pleura and have no direct association with large blood vessels and neither appeared to be supplied by blood vessels at all (Figure  4a ). Vascular pSALCs are associated with large blood vessels which they line directly (Figure  4b ). Adipose pSALCs are clusters of lymphocytes that reside on the surface of adipose tissue (Figure  4c , pointed out by yellow arrows). Protruding pSALCs have a three‐dimensional configuration in which they protrude, mainly as solitary structures, into the pleural cavity (Figure  4d , yellow dotted circle). Although most pSALCs could be assigned to one of these four categories, some showed characteristics fitting in with multiple categories, for example, pSALCs that reside on top of perivascular adipose tissue, therefore fitting in with both adipose and vascular pSALCs. Stereomicroscopic view of the four subtypes of pSALCs in hematoxylin‐stained resected pleural specimens. (a) Solitary pSALC. Blood vessels appear to be absent and SALCs appear to be flat structures. (b) Vascular pSALC. These SALCs line a blood vessel. (c) Adipose pSALC in the adipose streak. This type can be recognized as an area with increased cellular density and resides at the outer borders of the adipose tissue (yellow arrows). (d) Protruding pSALC. This subtype projects as a 3D structure into the pleural cavity (yellow oval). The left RPF (Figure  6a shows the left RPF in situ (orange dotted circle) in an opened thorax) contained various pSALC hot spots (Figure  5b , yellow and blue dotted ovals). The most densely populated hotspot was located close to the diaphragm and positioned between the adipose streak and the left phrenic nerve (Figure  5b , lower left dotted yellow oval). pSALCs in this location primarily presented as clusters of solitary pSALCs, although adipose SALCs were present as well. The next hotspot was immediately surrounding the left phrenic nerve (Figure  5b , blue dotted oval) which contained adipose pSALCs only (Figure  5c shows a magnification of the phrenic nerve, yellow arrows point out immune cells that reside on top of adipose tissue). The adipose streak, a distinct adipose structure within the RPF situated caudally to the heart (Figure  5b ), comprises another hotspot. This structure consistently harbored adipose pSALCs across all subjects, and occasionally protruding pSALCs. The next hotspot can be found in the RPF between the left phrenic nerve, the esophagus, and the diaphragm (Figure  5b lower right yellow dotted oval). Here, predominantly solitary pSALCs were observed, occasionally accompanied by small clusters of adipose pSALCs. Lastly, solitary pSALCs were occasionally detected caudally to the lung hilum (Figure  5b upper right yellow dotted circle). Stereo microscopic overview of the left RPF after hematoxylin staining. (a) Macroscopic view of the left RPF in situ. The orange dotted circle illustrates the approximate location of the tissue displayed in b. (b) Overview of the left RPF containing the adipose streak, esophagus and phrenic nerve (encircled by the blue oval). The yellow dotted ovals indicate SALC hotspots of the left RPF. (c) Magnified view of the left phrenic nerve lined with SALCs (yellow arrows). The right RPF (Figure  6a shows the right RPF in situ (orange dotted circle) in an opened thorax) contains various SALC hotspots as well (Figure  6b white, yellow and blue dotted ovals and circle). Consistently, adipose SALCs were observed in adipose tissue lining the inferior caval vein (Figure  6b , white dotted oval). The next hotspot could be allocated in the middle of the RPF (Figure  6b , yellow dotted oval) and was predominantly comprised of solitary SALCs. The most densely packed hotspot represented the caudal part of the right RPF (Figure  6b , blue dotted oval and circle) and contained mostly adipose SALCs that reside on adipose tissue lining vascular structures (Figure  6c shows a magnification of such an adipose SALCs), although occasionally some solitary SALCs and vascular SALCs could be observed. In contrast to the left phrenic nerve, the right phrenic nerve is surrounded by a substantial amount of adipose tissue. A few discrete immune cells were observed in this adipose tissue; however, definitive clusters were absent. Stereo microscopic overview of the right RPF after hematoxylin staining. (a) Macroscopic view of the right RPF in situ. The orange dotted circle illustrates the approximate location of the tissue displayed in b. (b) Overview of the right RPF containing the adipose streak, inferior caval vein and phrenic nerve. Dotted areas indicate pSALC hotspots. White: Adipose SALCs in adipose tissue lining the caval vein. Yellow: Solitary SALCs in the middle of the RPF. Blue: Adipose SALCs in the caudal part of the RPF (here SALCs primarily occurred in the oval area and rarely in the encircled area). (c) Magnified view of adipose SALCs. The yellow arrows indicate fine vascular structures that supply adipose tissue (white asterisks) with SALCs on their outer rim (red arrows). Pulmonary ligaments (Figure  7a shows the right pulmonary ligament in situ in an opened thorax (orange dotted oval)), proved to be very delicate, with an unfortunate instance of tearing in one rat during the resection procedure, rendering examination unfeasible. Nonetheless, in the five remaining rats, SALCs were consistently present within the pulmonary ligaments, irrespective of their side (left or right). Protruding SALCs were detected both along the portion of the ligament connecting the esophagus to the lungs (Figure  7b , yellow arrows), but also in the ligament between lung lobes (Figure  7c , yellow arrows). Stereo microscopic overview of the right pulmonary ligament after hematoxylin staining. (a) Macroscopic view of the right sided opened thorax. The orange dotted circle illustrates the approximate location of the tissue displayed in b. (b) Overview of the right pulmonary ligament (yellow dotted encircled structure) running from the esophagus to the hilum of the lung (lung is resected). The yellow arrows indicate protruding oval shaped SALCs. (c) Magnified view of the right pulmonary ligament between lung lobes. Yellow arrows indicate protruding SALCs. None of the DMPs contained SALCs (Figure  8a shows a resected hematoxylin stained DMP including its adjacent structures being the esophagus and subvertebral adipose tissue). A few DMP samples did contain some individual lymphocytes which seemed to be associated with fine vascular structures potentially representing lymphatics or blood vessels (Figure  8b , yellow arrows point out individual lymphocytes). All VMPs displayed adipose SALCs (Figure  8c , red arrows) and comparable to DMP occasionally contained individual lymphocytes (Figure  8c , yellow arrows). Stereo microscopic view of a resected and hematoxylin stained DMP and VMP. (a) Magnified view of the DMP, which contains no SALCs. (b) Magnified view of the DMP. (c) Magnified view of the VMP which contains adipose SALCs (red arrows). Yellow arrows: Single immune cells that often appear to have a relation with fine vascular structures. In summary, solitary and adipose pSALCs were the most common types, followed by protruding pSALCs. Vascular pSALCs were the least common type. Table  3 (third column) lists the occurrence of the different subtypes per pleural segment. Presence of T cells, B cells, and macrophages in different types of pSALCs. Note : The abundance of T and B cells and macrophages was evaluated by considering the relative number of fluorescent immune cells per pSALC section by using the following density scale −: complete absence, +: low density, ++: moderate density and +++: high density of fluorescent immune cells. The order of relative abundance from immune cells is listed from high to low in the most right column. Results are listed per rat. Some samples were divided in two, resulting in two levels of the same sample. Abbreviations: B, B cells; M, macrophages; T, T cells. Of each type of pSALC samples was resected and studied microscopically. Panel A of Figures  9 , 10 , 11 , 12 contains a stereomicroscopic view of a hematoxylin stained resected pSALC, whereas panel B of the same figures displays a corresponding microscopic HE stained section. HE sections clearly show the presence of dense clusters of immune cells in these structures, suggestive of SALC identity. Except for solitary pSALCs, all SALCs contained vascular structures (black arrows in panels A and B in Figures  9 , 10 , 11 , 12 ), with the vascular pSALC containing the most prominent vascular structures (Figure  10a,b , black arrows). Similar observations were made with respect to the presence of adipocytes (black asterisks in panel A and B in Figures  9 , 10 , 11 , 12 ). Solitary pSALCs appeared to be composed of immune cells only, whereas all other types contained adipocytes to some extent, with adipose pSALCs containing a clear core of adipocytes that are lined by a sheet of immune cells at its outer rim (Figure  11 , red arrows). Illustrative example of a solitary pSALC and its immune cell composition. (a) Stereomicroscopic view of two adjacent hematoxylin‐stained solitary pSALCs which appeared in a pleural segment as discrete lymphoid structures without any neighboring structures or clearly visible supplying vascular structures. (b) Bright field microscopic view of a HE‐stained tissue section of the pSALC in a, showing a high density of immune cells. No blood vessels were observed in this type of pSALCs. (c–e) Fluorescent microscopic view of sections stained with immune markers for T cells (CD3), macrophages (CD68) and B cells (CD79a). In general, it appears that T cells and macrophages are distributed more homogeneously, whereas B cells appear to show a tendency to occur at the outer rims of the SALC. Illustrative example of a vascular pSALC and its immune cell composition. (a) Stereomicroscopic view of a hematoxylin‐stained cluster of vascular pSALCs in a resected pleural segment. Vascular structures containing erythrocytes are clearly visible (arrows) and are lined by SALCs. In some clusters, adipocytes (black asterisks) could be discerned, thereby representing adipose pSALCs. (b) Bright field microscopic view of a HE‐stained tissue section of the pSALC in a, showing a high density of immune cells. Vascular structures filled with erythrocytes could be discerned (arrows). (c–e) Fluorescent microscopic view of sections stained with immune markers for T cells (CD3), macrophages (CD68) and B cells (CD79a). In general, it appears that T cells and macrophages are distributed more homogeneously, whereas B cells appear to show a tendency to occur at the outer rims of the SALC. Illustrative example of a cluster of protruding pSALCs and their immune cell composition. (a) Stereomicroscopic view of a hematoxylin‐stained cluster of protruding pSALCs (red arrows) that line adipose tissue (black asterisk) containing small vascular structures (black arrows). Protruding SALCs differ from other pSALCs by protruding into the pleural cavity. (b) Bright field microscopic view of a HE‐stained tissue section of the pSALCs in a, showing a high density of immune cells adjacent to adipocytes (black asterisks). Few small vascular structures could be discerned (black arrow). (c–e) Fluorescent microscopic view of sections stained with immune markers for T cells (CD3), macrophages (CD68) and B cells (CD79a). In general, it appears that T cells and macrophages are distributed more homogeneously, whereas B cells appear to show a tendency to occur at the outer rims of the SALC. All types of pSALCs contained T cells, macrophages, and B cells as shown in panels C–E of Figures  9 , 10 , 11 , 12 , respectively (CD3 = T cells, CD68 = macrophages and CD79a = B cells). Regardless of pSALC type, the topographical distribution of T cells and macrophages within pSALCs appeared diffuse, whereas B cells showed a tendency to be predominantly located at the periphery. After placing the relative contribution of immune cells in order from high to low (most right column in Table  3 ), it appears that B cells most often represent the highest cellular constituent and T cells most often represented the lowest constituent. The abundance of various immune cell types varied highly across pSALCs and no clear patterns of their ratio per pSALC type could be discerned. One adipose SALC contained a location with more pronounced organization that resembled a follicle as it showed a cluster of B cells surrounded by T cells (Figure  11c,e ). None of the other SALCs showed follicle‐like structures. Illustrative example of a large adipose pSALC and its immune cell composition. (a) Stereomicroscopic view of a hematoxylin stained adipose pSALC containing vascular structures (black arrows). Adipose pSALCs have a core of adipose tissue which is bordered by a sheet of immune cells on its outer rims (red arrows). (b) Bright field microscopic view of a HE stained tissue section of the pSALC in a, showing a sheet of immune cells adjacent to adipocytes (black asterisks). Few small vascular structures could be discerned (black arrows). (c–e) Fluorescent microscopic view of sections stained with immune markers for T cells (CD3), macrophages (CD68) and B cells (CD79a). In general, it appears that T cells and macrophages are distributed more homogeneously, whereas B cells appear to show a tendency to occur at the outer rims of the SALC. Interestingly, in one adipose SALC (the on displayed in this figure) it appears as if a follicle is developing as this specific area shows a cluster of B cells, surrounded by T cells (dotted grey circle points out this follicle in c and e).

The present study offers a thorough depiction, complemented by visual support, of the segmental anatomy of the rat parietal pleura and subsequently provides valuable insights into the distribution and morphological subtypes of pSALCs. This data will provide guidance for future studies of pSALCs in animals and humans. Most previous studies in rats refer to the parietal pleura as a whole and lack a clear differentiation between different segments, leading to ambiguity in referencing specific areas and previous topographical pSALC data (Cooray,  1949 ; Mixter,  1941 ; Pereira et al.,  1994 ). The current study, therefore, provides detailed morphological descriptions of the different pleural segments (Figure  1 ). This comprehensive approach hopefully enables future studies to align more effectively and provides guidance for future studies in human pleura, as extrapolation of rat data on pSALCs necessitates identifying corresponding sections of the parietal pleura in humans. Although most pleural segments in rats are present in humans, the DMP, VMP, and RPFs are unique to rats. In rats, the DMP is the result of a pleura fold that suspends the esophagus in the thoracic cavity, whereas in humans the esophagus is positioned directly on the vertebral column. The VMP and RPF are the result of a more cranially and dorsally positioned heart in rats when compared to humans. Given the anatomical positioning of the VMP, DMP, and RPF in the midline, it is reasonable to infer that these subdivisions in rats correspond to what is referred to as the mediastinal pleura in humans. In situ, SALCs appear as white cloudy spots, which are difficult to discern by the naked eye due to their relatively small size and low contrast to surrounding tissues. However, by means of hematoxylin staining of the opened thorax, pSALC visibility in situ and in subsequent resected pleural segments was significantly enhanced. This method allowed for a thorough in situ and three‐dimensional examination of pSALC distribution and morphology and allowed for targeted resection of individual pSALCs for additional microscopic studies (Schurink et al.,  2019 ). Additionally, it enabled the generation of photographs that provide readers with a clearer understanding of their anatomical configuration compared to previous, more descriptive studies. In the current study, pSALCs were observed in the left and right RPF (including the adipose streak), left and right pulmonary ligament, and the VMP, with the left and right RPFs exhibiting the highest density of pSALCs. The presence of pSALCs in these regions has been described previously (see Table  1 ) and hence their human counterparts should be included in future studies. These previous studies mention that pSALCs might occur in the pericardial, paravertebral, and diaphragmatic pleura as well, and although not observed in the current study, their corresponding pleura segments in humans could be of interest and should be included in future studies as well. Interestingly, it seems that in humans, at least in young individuals, the distribution of pSALCs may be more widespread, as lymphoid structures resembling pSALCs have been not only observed in pleural adipose tissue extending along the ribs from paravertebral to parasternal regions (Aharinejad et al.,  1990 ). Whether the differences observed between humans and rodents indicate interspecies variation or are due to the possibility that pSALCs may, as previously suggested, be transient in certain areas in younger individuals (Aharinejad et al.,  1990 ), warrants further investigation. Therefore, future studies in humans should not only focus on the pleural ligaments and mediastinal pleura but also include samples from the paravertebral, costal, and diaphragmatic pleura to fully comprehend the distribution of pSALCs in younger and adult human anatomy. The current study reveals that rats display four distinct subtypes of pSALCs: solitary, vascular, adipose, and protruding SALCs. To our knowledge, previous investigations on rat pSALCs have not made these distinctions (Cooray,  1949 ; Mixter,  1941 ; Pereira et al.,  1994 ), making our finding novel in this regard. Our observations on different types of pSALCs may stem from the fact that previous studies on pSALCs in rats primarily focused on microscopic sections, potentially overlooking various three‐dimensional characteristics. Interestingly, upon closer examination of the literature on morphological variances amongst peritoneal SALCs, a striking similarity was observed. Previous studies have reported four distinct types of peritoneal SALCs: milky spots, which can be either avascular or vascular (resembling solitary and vascular pSALC, respectively), fat‐associated lymphoid clusters (FALCs) (resembling adipose pSALCs), and foliate lymphoid aggregates (FLAGs) (resembling protruding pSALCs) (Bénézech et al.,  2015 ; Elewa et al.,  2014 ; Jia et al.,  2020 ; Moro et al.,  2010 ; Takemori et al.,  1994 ). Based on these observations, it can be assumed that morphological variance is a general SALC characteristic, irrespective of their associated body cavity. So far, no clear functional differences have been listed in the literature for the different morphological subtypes of peritoneal and pSALCs. Hypothetically, (1) different shapes might equal a different function, (2) different shapes might represent a manifestation of immunological activation, or (3) different shapes reflect different ontogenic developmental stages. Although previous studies on pSALCs do not show any difference between the various pSALCs with respect to their absorption and immune response after injection of substances, including tumor cells, immune cells, and ink particles (Mixter,  1941 ), it is, however, known that after activation, omental SALCs increase in size and subsequently show segregation of distinct B and T cell areas as well (Dux et al.,  1986 ). This implies that SALCs are dynamic structures not only with respect to their function but potentially also with respect to their shape. If SALCs increase in size after activation, their eventual shape, as we hypothesize, may be under the influence of the mechanical properties of their surrounding tissues. If surrounding tissues are pliable, SALCs can grow in all directions, resulting in a flat wide SALC, whereas rigid surrounding tissues might result in growth in one specific direction, resulting in a protruding SALC. A similar principle might apply to ontogenetically older SALCs, and hence more developed and larger in size when compared to younger SALCs. Future developmental and experimental studies, particularly those involving immune challenges, could offer valuable insights into this phenomenon. Microscopic evaluation of pSALC tissue sections showed that these structures are immediately positioned underneath the covering mesothelium, lack a capsule and trabeculae, and are primarily composed of T cells, B cells, and macrophages without a clear compartmentalized organisation. Since these are all SALC characteristics (Havrlentova et al.,  2017 ; Krist et al.,  1997 ; Shimotsuma et al.,  1991 ), it was concluded that all structures that were initially identified as SALCs in situ hematoxylin‐stained pleura indeed represented SALCs. Regardless of pSALC type, the distribution of T cells and macrophages appeared to be diffuse, whereas B cells showed the tendency to be located somewhat more at the periphery, an observation that fits in with earlier studies (Krist et al.,  1995 ; Liu et al.,  2015 ; Shimotsuma et al.,  1991 ). Interestingly, in one adipose SALC (see Figure  10 ) it appeared that there was a location with more pronounced organisation that resembled a follicle as it showed a cluster of B cells surrounded by T cells. None of the other SALCs showed follicle‐like structures. So far, FLAGs are the only peritoneal SALC type that is distinct from the other ones as it has been shown that they express partial T/B cell compartmentalization and that stromal cells in these distinct areas express different chemokines involved in lymphocyte homing (Jia et al.,  2020 ). It could then indeed be possible that FLAGs and potentially protruding pSALC are more developmentally mature or activated structures. In the current study, however, we did not observe lymphocyte compartmentalization in protruding SALCs. Nonetheless, stromal organization precedes lymphocyte organization, suggesting that the observed protruding pSALCs may represent an earlier developmental or activation stage. To deepen our comprehension of this phenomenon, forthcoming developmental and experimental studies could integrate an analysis of chemokine expression. This approach would afford valuable insights into the mechanisms underlying SALC development and function. Comparable to their peritoneal counterpart, experimental studies have shown that pSALCs absorb pleural fluid, induce an immune response, provide activated immune cells to the pleural cavity, and represent structures that prime specific B cells (Kuper et al.,  2021 ; Mixter,  1941 ; Pereira et al.,  1994 ). The latter will be released to the pleural cavity and hence translocate to the lung mucosa, contributing to local innate immune response by IgM production (Jackson‐Jones et al.,  2016 ; Weber et al.,  2014 ). These functions illustrate their relevance in innate and adaptive pleural, but also lung, immune homeostasis. From a toxicological perspective, pSALCs have gained increased interest since inhaled (nano) particles can be translocated over the visceral pleura to the pleural fluid, whereafter they can be resorbed in pSALCs and affect their function (Bénézech et al.,  2015 ; Bernstein et al.,  2015 ; Kuper et al.,  2018 ; Murphy et al.,  2011 ). In humans, lymphoid structures that could correspond to pSALCs have previously been identified exclusively in neonates and infants (Aharinejad et al.,  1990 ; Kampmeier,  1928 ; Mixter,  1941 ). However, preliminary findings from a cadaveric study suggest these structures may also be present in adults (Cleypool et al., unpublished data). If these structures do indeed represent pSALCs, their presence could have significant clinical implications. Inhaled substances, including toxins and drugs, could translocate from the parenchyma to the pleural fluid and subsequently get absorbed by pSALCs, potentially influencing drug pharmacokinetics and pSALC‐mediated immune responses. Furthermore, SALCs are thought to create a microenvironment conducive to tumor invasion and proliferation (Cruz‐Migoni & Caamaño,  2016 ; Litbarg et al.,  2007 ). Exploring pSALC–metastasis interactions could provide valuable insights into their role in cancer progression and their potential as targets for onco‐immunotherapy. Additionally, the pSALC microenvironment may facilitate endometrial cell growth as these cells exhibit cancer‐like properties, including resistance to apoptosis, uncontrolled proliferation, invasion, and local immune suppression (Moghaddam et al.,  2022 ). This raises the possibility that pSALCs play a role in the pathogenesis of pleural endometriosis. Furthermore, given the fact that sympathetic nerves have been identified in proximity to immune cells in other human secondary lymphoid structures including spleen, lymph nodes, and omental SALCs (Cleypool et al.,  2019 ; Cleypool, Brinkman, et al.,  2021 ; Cleypool, Mackaaij, et al.,  2021 ), it is plausible that pSALCs are innervated by the sympathetic nervous system as well. This raises the possibility of neuroimmune interactions within the pleural microenvironment which potentially could represent a novel therapeutic target for the treatment of various lung and pleura‐associated inflammatory diseases. These highlight the potential roles of pSALCs in various diseases and conditions underscoring the necessity for further exploration of these structures in humans.

In conclusion, various segments of the rat parietal pleura contain lymphoid structures that identify as SALCs and exhibit diverse shapes akin to their peritoneal counterparts. This dataset advances our comprehension of pSALCs in rats and provides valuable insights for designing future experimental investigations and studies aimed at uncovering analogous lymphoid structures within the human pleura. If morphologically identical structures are found in the human parietal pleura, it could imply analogous functions as well. Such findings will have the potential to reshape our understanding of certain lung and pleura‐associated diseases and may contribute to the development of preventive, diagnostic, and therapeutic strategies.

The lymphoid system comprises various lymphoid structures, including the spleen, which monitors blood, lymph nodes, monitoring lymph fluid, and the less well‐known serosa‐associated lymphoid clusters (SALCs; including fat‐associated lymphoid clusters (FALCs) and milky spots), responsible for monitoring body cavity fluid (Van Vugt et al.,  1996 ; Wilkosz et al.,  2005 ). SALCs are primarily composed of macrophages, B cells, and T cells and, in contrast to the spleen and lymph nodes, lack a clearly defined cellular organization and capsule (Cleypool et al.,  2020 ). The functions of SALCs can be categorized into three main aspects: (1) they absorb fluid from their associated body cavity through stomata of the mesothelium, monitoring it for foreign or pathogenic substances (Meza‐Perez & Randall,  2017 ; Van Vugt et al.,  1996 ; Wilkosz et al.,  2005 ), (2) they serve as a conduit for circulating immune cells recruited to the body cavity during a serosal immune challenge (Wijffels et al.,  1992 ), and (3) they represent a structure in which a specific B cell type, B1 or CD5 B cells, develops whereafter they, via the body cavity, translocate to the intestinal or pleural mucosa and provide an IgM baseline (Platell et al.,  2000 ; Rothstein et al.,  2013 ). Furthermore, SALCs are known to represent hotbeds for the development of serosa associated metastatic disease; exfoliated tumor cells from a primary tumor can disseminate to peritoneal or pleural fluid, get absorbed by SALCs, where they can develop into secondary tumors (Hagiwara et al.,  1993 ; Tsujimoto et al.,  1995 ) due to an optimal tumor microenvironment in these SALCs (Gerber et al.,  2006 ; Ladanyi et al.,  2018 ). In humans, SALCs have predominantly been studied in the greater omentum, where they are also referred to as milky spots. If similar structures, with analogous features, are present in human pleura as well (from now on referred to as pleural (p) SALCs) it is of great significance to gain a better understanding, as their existence has profound implications for comprehending pleura and lung‐associated diseases and for future therapy development. So far, in humans, lymphoid structures that could represent pSALCs have been studied in neonates and infants only (Aharinejad et al.,  1990 ; Kampmeier,  1928 ; Mixter,  1941 ). Although a pilot study suggests they might be present in adults as well (Cleypool et al., unpublished data), currently thorough data is lacking. In order to consider extrapolation of functional characteristics of pSALCs from rodents to humans, rodent pSALCs and the observed lymphoid cell clusters in human parietal pleura should at least show morphological similarities. For this interspecies comparison, it is of significance to have a proper understanding of rodent pSALCs with respect to their in situ anatomical configuration (e.g. location, number, and cellular composition). Data on pSALC function derives mainly from toxicological studies in rats. However, these rat pSALCs studies show images of microscopic sections only (Cooray,  1949 ; Mixter,  1941 ; Pereira et al.,  1994 ) which do not allow one to fully understand and hence compare their anatomy to observations in humans. Furthermore, these studies are inconsistent when it comes to naming various parts of the parietal pleura, thereby making it difficult to fully interpret the presented topographic data. To overcome these issues and to be able to extrapolate experimental pSALC data from rats to humans, this anatomical study in rats aims to (1) establish a proper description (with visual support) of various parts of the parietal pleura, (2) extract topographic data on pSALCs in situ, (3) discern various types of pSALCs, and (4) determine pSALC cellular composition.

Literature was studied and all previously observed or suggested pSALC locations in rats were extracted (Table  1 contains an overview and description of these locations). pSALCs have been most frequently observed in retrocardiac pleural folds (RPFs) (Cooray,  1949 ; Mixter,  1941 ; Pereira et al.,  1994 ) and the pleural fold in between the esophagus and the aorta (Cooray,  1949 ; Mixter,  1941 ). Other mentioned or indicated pSALC locations are the dorsal and ventral mediastinal pleura (DMP and VMP, respectively), the pericardial pleura, the paravertebral pleura, the pulmonary ligaments, and the diaphragmatic pleura (Mixter,  1941 ). Since these older studies not always give clear descriptions of these locations and do not contain illustrative images, interpretation of data as presented in these older studies might be somewhat arbitrary. Observed and suggested rat pSALC locations described in the literature. Rat cadavers were fixed and dissected to study the above‐mentioned locations, both to get a better understanding of the anatomy of the different parts of the parietal pleura and for the presence and morphological characteristics of pSALCs in situ. To confirm their SALC identity, various pSALCs were resected and microscopically evaluated for the presence of lymphocytes and macrophages, the primary constituents of SALCs (Krist et al.,  1997 ; Liu et al.,  2015 ; Shimotsuma et al.,  1991 ). Six surplus wildtype Long Evans female rats (5–6 months of age) were included in this study. These rats were previously used for non‐invasive experiments, whereafter the animals were euthanized by carbon dioxide inhalation. The use of surplus cadaveric tissues did not require ethics committee approval. After euthanasia, all abdomens were opened by a midline incision, and the pleural cavities were injected with 4% formaldehyde through the diaphragm. Rats were further fixed by immersion in 4% formaldehyde for 1 week, subsequently rinsed in running tap water for 24 h, and stored at 4°C in a 0.1 M phosphate buffer containing 15% sucrose (pH 7.4) until further use. The supplementary data contains a step‐by‐step dissection guide describing how to isolate the thorax, locate the pleura, and how to resect all relevant structures for further studies. After opening of the fixed thorax, all segments of the pleura, which include the left and right RPF, left and right pericardial pleura, left and right costal pleura, VMP, DMP, diaphragmatic pleura, and the left and right paravertebral pleura, were located and images of their in situ anatomy were recorded. These images were used to construct a proper understanding of the different segments of the pleura (Figure  1 ). Anatomy and nomenclature of thoracic structures including the various segments of the parietal pleura. Top row: Medial view of the opened right side of the thorax. The rib cage and right lung are removed to gain a better view of the various parts of the parietal pleura. Bottom row: Medial view of the opened left side of the thorax. The VMP is not visible in these views. V: Ventral side of the thorax. D: Dorsal side of the thorax. Since pSALCs were difficult to observe with the naked eye, opened thoraxes were studied with the aid of a stereomicroscope (Leica EZ4, Nussloch, Germany). In situ, SALCs appear as white cloudy spots, which were difficult to discern by the naked eye due to their relatively small size and low contrast to surrounding tissues. By means of hematoxylin staining of the thoracic cavity, pSALC visibility in situ was significantly enhanced (see supplementary data for details on this staining method). In some pleural segments, SALCs were still difficult to discern since stained SALCs showed too little contrast with respect to the background, especially if these pleural segments were on top of an organ or folded. In these cases, the pleural segment was resected and placed in a petri dish, increasing pSALC visibility significantly. All pleural segments were studied for the presence of pSALCs, and their relative abundance per segment was evaluated by considering the area of hematoxylin‐stained lymphoid tissue by using the following density scale −: complete absence, +: low amount, ++: moderate amount, and +++: high amount. General notes on morphological characteristics of pSALCs were recorded. After evaluation of these notes, it became clear that four types of pSALCs could be discerned (the different types are described in detail in the “Results” section). Macroscopic photographs were obtained using a mobile phone camera (Figure  1 ). Stereomicroscopic images were obtained with a mobile phone camera through the objective of the stereomicroscope, resulting in circular images (Figures  2 , 3 , 4 , 5 , 6 , 7 , 8 , 9 ). Stereomicroscopic view of a right‐sided ventral mediastinal pleura (VMP) in situ in a hematoxylin‐stained thorax. After carefully pulling the sternum ventrally, the VMP becomes visible between the heart and the sternum. Of each of the four pSALC types, multiple samples were resected and separately placed in a wells plate filled with 0.1 M phosphate buffer, containing 15% sucrose (pH 7.4) and stored at 4°C until further processing for paraffin embedding. Table  3 contains an overview of the resected pSALCs. Due to a significant time of storage, the hematoxylin staining had become weaker, and to increase pSALC visibility during the dehydration process, paraffin embedding, and cutting, all samples were re‐stained with hematoxylin. This staining was performed by immersion of samples for 5 min in hematoxylin (by removing the phosphate buffer from the wells and by adding hematoxylin) and subsequent rinsing in tap water (remove hematoxylin from wells and apply clean tap water for three times). All samples were then placed in small steel containers and further processed by dehydration through increasing concentrations of ethanol, clearing them in xylene (these steps should be no longer than 45 min to prevent wash out of hematoxylin) and embedded in paraffin. Sections were cut at 5 μm using a microtome (Autocut R Histocore, Leica, Nussloch, Germany), placed on glass slides, dried, and heat fixed by subsequently placing the slides on a drying table (Medax, 14801, Kiel, Germany) at 60°C for 1 h, followed by overnight heat fixation in an incubation oven at 60°C (Binder, Tuttlingen, Germany). All sections were deparaffinized and rehydrated prior to hematoxylin and eosin (HE) staining and immunohistochemical staining. HE was used to generate a general tissue overview, and immunohistochemical staining was performed on adjacent sections with antibodies raised against CD3, CD79a, and CD68 proteins, thereby visualizing T cells, B cells, and macrophages, respectively. Tissue sections were stained with hematoxylin (Klinipath, Olden, Belgium) for 10 min. After rinsing in running tap water, sections were dipped in ethanol 50%, stained with eosin (Klinipath) for 1 min, and dehydrated in a graded series of alcohol and xylene. The slides were cover slipped with Entellan (Merck, Darmstadt, Germany). All steps were performed at room temperature (RT). Tissue sections were pretreated with heat induced epitope retrieval by placing sections in citrate buffer (0,1 M, pH 6.0) for 20 min at 95°C. Tris‐buffered saline with 0.05% Tween20 (TBS‐T) was used for all washing steps and was performed at RT. All sections were washed with TBS/T and pre‐incubated for 10 min with 3% normal rat serum in TBS‐T, followed by incubation with primary antibodies. Rabbit anti‐human CD3 antibody (1:50 in TBS‐T + 1% BSA, 90 min, RT, Dako, Glostrup, Denmark), mouse anti‐human CD68 antibody (1:250 in TBS‐T + 3% BSA, overnight at 4°C, Dako) or Rabbit anti‐CD79a (1:1000 in TBS‐T + 3% BSA, overnight at 4°C, Abclonal, Woburn MA, United States of America). After washing in TBS‐T, the sections were incubated with either undiluted Brightvision Poly‐AP Goat‐anti‐Rabbit (ImmunoLogic, Amsterdam, the Netherlands) or undiluted Brightvision Poly‐AP Goat‐anti‐Mouse (ImmunoLogic) for 30 min at RT. Immunoreactivity was visualized with liquid permanent red (LPR) (Dako). Sections were counterstained with hematoxylin (Klinipath), dried on a hotplate for 15 min at 60°C, and cover slipped with Entellan (Merck). Sections of rat spleen were used as positive controls for all markers. Negative controls were obtained by incubation of pSALC sections with TBS‐3% BSA without primary antibodies. All slides were evaluated by bright field microscopy and in case of immunohistochemical staining by additional fluorescent microscopy (the chromogen LPR has stable fluorescent features, allowing to alternately switch between bright field and fluorescent microscopy). All samples were studied using a DM6 microscope (Leica, Nussloch, Germany) and in case of fluorescent microscopy, an I3 fluorescent filter. The relative abundance of T and B cells and macrophages was evaluated by considering the number of fluorescent immune cells by using the following density scale −: complete absence, +: low density, ++: moderate density, and +++: high density of fluorescent immune cells. If a specific distribution pattern was observed, this was noted as well. Each sample was examined independently by two observers (Mackaaij and Cleypool) who were masked for the sample origin. When there was disagreement between the observers, the observed samples were re‐examined by the same observers and scored by consensus. If a specific distribution pattern was observed, this was noted as well. Brightfield and fluorescent images were captured at various magnifications using a DM6 microscope with a motorized scanning stage, a DFC7000 T camera, and LASX software (all from Leica, Nussloch, Germany).

Data S1.