Afferent Lymph Research Articles

Unilateral smoke inhalation in sheep: effect on left lung lymph flow with right lung injury.

We previously reported that smoke inhalation to the right lung will result in damage to the air-insufflated left lung. In this study we confirm these findings and determine whether this injury is associated with an elevation in lung lymph flow and pulmonary microvascular permeability to protein as indexed by changes in reflection coefficient. Sheep (n = 12) were surgically prepared by placement of a Swan-Ganz catheter and pneumatic occluders on all pulmonary veins and the left pulmonary artery. The left lung lymphatic was selectively cannulated as shown previously (Y. Kikuchi, H. Nakazawa, and D. Traber. Am. J. Physiol. 269 (Regulatory Integrative Comp. Physiol. 38): R943-R947, 1995). All afferent lymphatics from the right lung were severed, and the right pulmonary ligament was sectioned. The caudal end of the lymph node was sectioned to remove systemic lymph contamination. The sheep were studied in the unanesthetized state 7 days later. To ensure that lymph flow was exclusively from the left lung (QLL), right pulmonary microvascular pressure was increased, a procedure that resulted in little or no change in QLL, as was previously shown. The sheep were then anesthetized, and a Carlens tube was positioned to allow separate ventilation of the right and left lung. The right lungs of five sheep and the left lungs of two sheep were insufflated with cotton smoke. Insufflation of the left lung with cotton smoke produced a fourfold increase in QLL that began 4 h after insult. Insufflation of the right lung with smoke led to a doubling of QLL that began 12 h after insult. Changes in QLL were associated with increased microvascular permeability, as indexed by the reflection coefficient. Control sheep (air insufflated into both lungs, n = 5) showed no change in QLL. Injury to the right lung resulted in damage to the left air-insufflated lung, suggesting a hematogenous mediation of the response.

Local cell traffic and cytokine production associated with ectoparasite infection

This paper reviews recent advances in our understanding of changes in local cellular traffic and cytokine synthesis that occur as a result of infection of sheep with the ectoparasite Lucilia cuprina. Changes in the cellular composition and cytokine profile of infected skin and draining afferent and efferent lymph were assessed using standard approaches and, in addition, a variety of techniques that have only recently become available as a result of advances in ruminant cytokine biology. These include cytokine-specific immunoassay, reverse transcription PCR (RT-PCR) and immunohistology. The initial acute inflammatory response was characterised by the infiltration of polymorphonuclear cells followed by selected lymphocyte subsets into discrete areas adjacent to the site of infection. Analysis of cytokine expression in skin prior to and following infection provided a molecular basis for the observed cellular events. Both cellular and molecular events within the skin were reflected within draining afferent lymph providing a basis for the conclusion that events within the skin (other than antigen uptake and transport) may influence events within the draining node and thus the outcome of the immune response to the parasite. Analysis of cellular and molecular changes in efferent lymph during infection suggested initiation of antigen-specific immunity.

Cytokines and their inhibitors in orf virus infection

The epitheliotropic parapoxvirus, orf virus, can repeatedly infect sheep skin. A specific immune response is generated as reinfections induce smaller lesions with quicker resolution times than primary lesions. Cyclosporin-A treatment abrogates this partial immunity. Cytokine mRNAs detected in lesion biopsies include the transcripts for IL-1β, IL-3 GM-CSF, TNF-α and, less reproducibly, IFN-γ. CD4 + T-cells predominate in afferent lymph draining the site of infection, and are the major source of GM-CSF and IFN-γ. IL-1β and IL-8 are also detected. The orf virus genome contains a homologue of mammalian vascular endothelial growth factor that may enhance virulence and a vaccinia virus E3L-like gene which may inhibit the anti-viral effect of the interferons. A GM-CSF inhibitory activity has also been discovered and has been ‘chased’ into a 10 kb DNA segment of the orf virus genome. These studies indicate that orf virus may temporarily avoid host immunity by a combination of acute, rapid infection and replication in the epidermis and by producing virulence factors that inhibit protective proteins of the host immune and inflammatory response.

Reactivity of antibodies directed against human antigens with surface markers on canine leukocytes.

A panel of anti-human antibodies was used in order to investigate their cross-reactivity with canine leukocytes. The labeling was carried out at the microscopic level by immunocytochemical staining. Of 50 antibodies 22 cross-reacted with canine leukocytes from afferent lymph and peripheral blood. All leukocytes reacted with MHM23 (CD18). Two anti-HLA DR antibodies, DK22 and L243, reacted mostly with veiled cells and with PHA-stimulated lymphocytes, whereas TAL1B5 was cross-reactive with all canine lymphocytes. The activation markers CD25, Ki-67, PCNA were identified on PHA-stimulated lymphocytes with ACT1, Ki-67 and PC10 clone produced antibodies. Canine eosinophils reacted with MHM6 (CD23) antibody. A large number of antibodies reacted with canine lymph veiled cells. Canine granulocytes and a subset of lymphocytes were stained with the anti-CD15 antibody only after treatment of cytospins with neuraminidase.

GM-CSF mRNA and protein in human skin-derived lymph.

GM-CSF together with IL-1 beta and TNF-alpha has been shown to play a key role in the maturation of LC in vitro. To investigate the presence of GM-CSF, IL-1 beta and TNF-alpha in human skin-derived lymph, we cannulated microsurgically a superficial lymph vessel on the lower leg of six healthy volunteers. Messenger RNA levels were estimated by a reverse transcriptase polymerase chain reaction (RT-PCR) method. From a total of 20 different samples, each consisting of 10(6) lymph cells, total RNA was extracted, reverse transcribed to cDNA and amplified using specific primers for the target gene. Amplified products were sized by electrophoresis and visualized by ethidium bromide. Specific transcripts for GM-CSF were detected in all lymph samples, indicating that circulating human skin-derived lymph cells express GM-CSF mRNA. A mean level of 11.5 +/- 2.1 pg/ml GM-CSF was detected in the lymph samples examined, as determined by a sensitive ELISA. In contrast to GM-CSF, occasional weak mRNA signals together with a mean level of 2.7 +/- 2.2 pg/ml were found for IL-1 beta, and neither specific transcripts nor protein were detected for TNF-alpha. Thus, our results demonstrate that afferent skin lymph cells constitutively express GM-CSF.

Rates of flow of technetium 99m-labeled human serum albumin from peripheral injection sites to sentinel lymph nodes

Background: The new technique of sentinel lymphadenectomy for cutaneous melanoma provided us with a unique opportunity to quantitate the rates of lymphatic flow in afferent lymphatics.

Lyphocyte migration in L-selectin-deficient mice. Altered subset migration and aging of the immune system.

Lymphocyte trafficking across high endothelial venules (HEV) of peripheral lymph nodes (PLN) is dependent upon lymphocyte expression of L-selectin. Mice that lack this adhesion molecule provide an opportunity to determine the long-term role of L-selectin-mediated migration in the maintenance of leukocyte subpopulations. HEV in L-selectin-deficient mice were phenotypically, morphologically, and functionally comparable with wild-type mice, although there was a 70 to 90% reduction in the number of lymphocytes within PLN. These lymphocytes most likely entered PLN through the afferent lymphatics, since they did not migrate into PLN of normal mice during short-term homing experiments. The impaired trafficking of lymphocytes across PLN-HEV resulted in the accumulation of memory (CD18highCD44high) lymphocytes within PLN, and also altered the distribution of lymphocyte subpopulations within other tissues. Specifically, a 30 to 55% increase in splenic cellularity occurred due to increases in both naive and memory lymphocytes. Circulating lymphocyte numbers or subpopulations were not altered in young L-selectin-deficient mice, but circulating monocyte numbers were increased nearly threefold. In contrast, older L-selectin-deficient mice had disproportionate increases of both naive and memory CD4+ T cells present within spleen and blood. These results and the finding that memory lymphocytes in wild-type mice expressed L-selectin demonstrate a requirement for L-selectin in the regulation of memory lymphocyte migration. Therefore, L-selectin-dependent pathways of lymphocyte migration are important for the normal migration of both naive and memory lymphocytes.

Afferent lymph veiled cells stimulate proliferative responses in allogeneic CD4+ and CD8+ T cells but not gamma delta TCR+ T cells.

Dendritic cells were identified in afferent lymph derived by lymphatic cannulation of cattle, stained with monoclonal antibody (mAb) to the bovine workshop cluster 6 (WC6) antigen, which is highly expressed on bovine afferent lymph veiled cells, and sorted with a fluorescence-activated cell sorter. These cells expressed major histocompatibility complex (MHC) class I and II and CD1b but not CD14. They bound human and murine CTLA4-immunoglobulin (CTLA4-Ig) fusion proteins indicating expression of CD80 and or CD86. Dendritic cells induced proliferative responses in allogeneic CD4+ and CD8+ cells sorted from blood but did not induce responses in purified allogeneic WC1+, gamma/delta T cells, which are CD2-, CD4-, CD8- and are the major gamma delta T-cell population in cattle blood, even when interleukin-2 (IL-2) was added to cultures. A WC1-, CD2+ gamma delta T-cell receptor (TCR)+ population predominates in cattle spleens and proliferation of a T-cell line with this phenotype was not induced by allogeneic dendritic cells, with or without added IL-2. The observations imply that the ligand for the gamma delta TCR expressed on the two populations is not present on allogeneic dendritic cells or that the costimulatory molecules expressed on dendritic cells that render them highly effective at stimulating MHC class I- and class II-restricted CD8+ and CD4+ T cells are not recognized by the WC1+ or WC1- gamma/delta T cells. Expression of CD28 by the four cell types was assessed by reverse transcriptase-polymerase chain reaction (RT-PCR). Purified CD4+ and CD8+ cells both produced CD28 transcripts but neither purified WC1+ cells nor the WC1- gamma delta TCR+ cell line did so. The findings indicate that CD80 and or CD86 are involved in the stimulation of CD4+ and CD8+ alpha beta TCR+ T cells but not in the stimulation of either of the two gamma delta TCR+ populations.

Rates of flow of technetium 99m--labeled human serum albumin from peripheral injection sites to sentinel lymph nodes.

The new technique of sentinel lymphadenectomy for cutaneous melanoma provided us with a unique opportunity to quantitate the rates of lymphatic flow in afferent lymphatics. Seventeen melanoma patients underwent preoperative lymphoscintigraphy with technetium 99m-human serum albumin (HSA). The time and distance between the injection site and the sentinel lymph node (LN) were recorded. By comparison, lymphatic flow rates between footpad, popliteal LN, femoral LN, and systemic blood were measured in 60 female mice (C57BL/6) after footpad injection of 99mTc-HSA. The rate of lymphatic flow to 14 axillary, four inguinal, one popliteal, and one parotid sentinel LNs averaged 10.4 +/- 7.3 cm/min. In contrast, the lymphatic flow rate between the footpad and the popliteal LN in mice (analogous to the sentinel LN in human beings) averaged 1.33 +/- 0.52 cm/min. There was a marked delay in the passage of radionuclide through the popliteal LN with consequent slowing of the rate of flow between the popliteal and femoral LNs to 0.22 cm/min. Lymphatic flow to the sentinel LN occurs rapidly from both human skin and murine footpads. This information might be helpful in planning the timing of the incision after vital blue dye injection for identifying the sentinel LN.

Cytolytic activity and associated serine protease expression by skin and afferent lymph CD8+ T cells during orf virus reinfection

Following orf virus reinfection in sheep, CD8+ T cells were recruited to the lesion site in the skin during the period of viral replication and appeared in increased numbers in afferent lymph draining the infection site. A proportion of the CD8+ T cells were activated, particularly in the skin, as determined by their expression of the cytolytic cell-associated serine protease, BLT-esterase. This was a useful marker for activated cytolytic cells as there was a good correlation between afferent lymph CD8+ T cell-mediated lysis of autologous orf virus-infected skin cells and their expression of BLT-esterase in vitro. The majority of the cells that expressed BLT-esterase in the skin and lymph were CD8+ T cells and cytolysis in vitro was predominately by the MHC class I antigen presentation pathway. The conclusion is that orf virus reinfection is characterized by limited viral replication and stimulation of an activated CTL response in the skin.

Identification of the sheep homologue of the monocyte cell surface molecule — CD14

An ovine monocyte/macrophage cell surface antigen was recognized by three mouse monoclonal antibodies (mAbs) VPM65, VPM66 and VPM67. These mAbs also reacted with bovine cells. The antibodies immunoprecipitated a single, glycosyl-phosphatidylinositol-linked polypeptide of M r 55 000 which, when deglycosylated, was reduced to M r 53 000. They reacted strongly with peripheral blood monocytes, alveolar macrophages and peripheral blood granulocytes, and weakly with afferent lymph dendritic cells. They also reacted with macrophages in many different tissues but were non-reactive with lymphocytes. Competitive flow cytometry shows that these three mAbs recognize the same or a closely related epitope of a single antigen. An antigen-specific capture ELISA using the anti-human CD14 mAb (TÜK4) revealed that all four mAbs associate with the same antigen. These data demonstrate that the mAbs react with the ovine homologue of the lipopolysaccharide (LPS)-LPS binding protein receptor, CD14.

Constitutive and inducible expression of interleukin‐6 by Langerhans cells and lymph node dendritic cells

During the induction phase of contact sensitization and other cutaneous immune responses a proportion of epidermal Langerhans cells (LC) is induced to leave the skin and migrate via afferent lymphatics to lymph nodes draining the site of exposure. The cells that accumulate in draining nodes have acquired the characteristics of immunostimulatory dendritic cells and effectively present antigen to responsive T lymphocytes. In the present study we have questioned whether LC in the epidermis and the lymph node dendritic cells into which they develop express interleukin-6 (IL-6), a cytokine that has been shown to serve as an important costimulator of T lymphocyte activation. In situ immunocytochemical analyses using a biotin-streptavidin staining technique revealed that dendritic cells resident in the epidermis of untreated mice constitutively express this cytokine. Keratinocytes expressed detectable IL-6 only following local exposure to the contact allergen oxazolone. Such treatment also appeared to enhance the expression by epidermal dendritic cells of this cytokine. Analyses of unfractionated and LC-enriched and -depleted populations of epidermal cells revealed a close correlation between major histocompatibility complex (MHC) class II (Ia) antigen expression and staining for IL-6, implicating LC as the sole or major source of this cytokine in unstimulated epidermis. Finally, compared with tissue isolated from mice treated with vehicle alone, draining lymph nodes prepared from animals 18 hr following sensitization with oxazolone displayed a substantial increase in both the frequency of dendritic cells and the number of IL-6+ cells within the paracortex. These data demonstrate that resident epidermal LC and the dendritic cells into which they develop are important sources of IL-6. Their constitutive and inducible expression of this cytokine will facilitate the induction of cutaneous immune responses.

The cytokine response of afferent lymph following orf virus reinfection of sheep.

The cytokine response of afferent lymph following orf virus reinfection of sheep.

Trypanosoma congolense infection in sheep: Cellular phenotypes in lymph and lymph nodes associated with skin reactions

Intradermal inoculation of sheep with culture-derived metacyclic forms of Trypanosoma congolense resulted in the development of localized skin reactions (chancres) and enlargement of the draining lymph nodes 7 days after infection. Changes in the expression of surface antigens of lymphocytes in lymph leaving the affected skin reactions and in the associated lymph nodes were monitored by cannulating the afferent and efferent lymphatic ducts. Trypanosomes appeared in afferent and efferent lymph 3 to 5 days after infection and persisted even as the chancres regressed. The cellular output in both afferent and efferent lymph increased markedly after the onset of parasitosis. Sequential analysis of the phenotypes of lymphocytes by immunofluorescent staining and flow cytometry revealed that in afferent lymph draining the chancre there was an early response which was due to an increase in T cells, particularly CD4+ and CD8+ cells; however, as the chancres-regressed there was an increase in lymphoblasts and surface immunoglobulin-bearing cells. In contrast, in the efferent lymph, the increase in lymphocytes was due predominantly to a higher number of cells bearing surface immunoglobulins.

Investigation of skin-derived lymph: an approach to get insight into the skin immune system.

The skin acts as a mechanical, physicochemical and immunological control and defense system, and its lymphatic vessels play a role in the regulation of cell hydration and osmosis, as well as in immunological responses. Skin-derived signals, produced in response to various agents acting on the skin, arrive with the afferent lymph at the regional lymph nodes. These signals reflect the immunological processes in the skin and may also determine the reactions in the lymph node. By analyzing afferent lymph derived from specific skin lesions it should therefore be possible to get insight into local pathomechanisms as well as the signal transmission in skin disorders. In this report, we present methodological aspects of such skin-derived lymph studies and review the results obtained from studies of sodium-lauryl-sulfate-induced irritative contact dermatitis in humans.

Enhanced proliferation of CD4 + T cells induced by dendritic cells following antigen uptake in the presence of specific antibody

Afferent lymph dendritic cells bear an Fcγ receptor which binds antigen/antibody complexes thereby enhancing uptake of antigen. In this report, we have addressed the question of whether the enhanced uptake of antigen results in augmented antigen presentation and T cell proliferation in in vitro secondary responses in sheep. Inclusion of affinity-purified IgG anti-ovalbumin antibody in cultures of afferent lymph dendritic cells, purified CD4 + T cells, and substimulating amounts of ovalbumin resulted in a five- to 169-fold enhancement of T cell proliferation. This effect was antigen-specific as replacement of the anti-ovalbumin antibody with an IgG anti-human serum albumin specific antibody did not cause enhanced T cell responses. The antigen-specific augmentation required intact antibody Fc portions as F(ab′) 2 fragments of the anti-ovalbumin antibodies were ineffective. The enhanced antigen presentation was found to be maximal with immune complexes in moderate antibody excess (three- to 30-fold), but still occurred at antibody/antigen ratios of 300. The augmented responses were inhibitable with anti-MHC Class II specific antibodies, indicating that at least some of the antigen taken in via Fcγ receptors entered a Class II processing pathway. The results thus show that antigen uptake via Fcγ receptors on dendritic cells results in functional augmentation of antigen presentation and T cell proliferation.

Fcγ Receptor Expression on Sheep Afferent Lymph Dendritic Cells and Rapid Modulation of Cell Surface Phenotype Following Fcγ Receptor Engagement In Vitro and In Vivo

Afferent lymph dendritic cells were analysed for the presence of Fc gamma receptors by Western blotting and for modulation of surface markers following Fc gamma receptor engagement in vitro and in vivo. The results showed that unstimulated dendritic cells expressed Fc gamma RII constitutively. When dendritic cells were incubated in vitro with antigen/antibody complexes in antibody excess, a marked reduction in surface staining was observed for MHC class II, CD1, CD44, and VLA-4 after 8 h in culture. These changes did not occur with antigen or antibody alone. DC expression of LFA-1 and LFA-3 were slightly reduced after 8 h in culture with Ova alone, but this was enhanced slightly when the cells were cultured with immune complexes. Even more marked reductions in surface staining for MHC class II, CD1, CD44 and VLA-4 were observed on dendritic cells 4-8 h following secondary antigen challenge in vivo. LFA-1 and LFA-3 expression was reduced only slightly. The level of expression of MHC class II, CD1, LFA-1 and LFA-3 was substantially increased over resting values 24 h after Fc gamma R occupancy. The intensity of staining at this time was also significantly elevated for CD44, LFA-1, LFA-3 and VLA-4. These results show that engagement of Fc gamma receptors cause a substantial modulation of the dendritic cell surface phenotype after immune complex uptake. The phenomenon may function to maximize subsequent presentation of the challenge antigen to T cells.

Low levels of HIV-1 infection in cutaneous dendritic cells promote extensive viral replication upon binding to memory CD4+ T cells.

Earlier work has identified a cell population that replicates HIV-1 in the absence of standard T cell stimuli. The system consists of dendritic cells and memory T lymphocytes that emigrate from organ cultures of human skin and together support a productive infection with HIV-1. These emigrants resemble cells that can be found in mucous membranes and that normally traffic in afferent lymph. Here, we report that a low level of infection in the dendritic cell can initiate extensive HIV-1 replication in cocultures with T cells. First we extended our earlier work to larger skin specimens from cadavers. As long as the organ cultures were set up within 36 h of death, the emigrant leukocytes were comparable to cells from fresh surgical specimens in number, phenotype, and function. These mixtures of dendritic cells and T cells provided the milieu for a productive infection with several virus isolates. When purified dendritic cells were separately pulsed with virus and then mixed with T cells that had not been pulsed with HIV-1, active infection ensued. The infectivity of HIV-pulsed dendritic cells persisted for at least 1.5 d in culture, but was blocked if AZT was added during that time to block reverse transcription in the dendritic cells. The number of copies of proviral DNA in the dendritic cells corresponded to < 100 copies per 5 X 10(4) cells, but upon mixing with T cells, > 10(4) copies were found 5-7 d later. By contacting syngeneic T cells, extralymphoid depots of dendritic cells--even with a low viral burden as has been reported in vivo--may contribute to chronic HIV-1 replication in infected individuals.

The role of selective lymphadenectomy in the management of patients with malignant melanoma.

A novel surgical technique based on selective lymphadenectomy was used to stage 132 patients with intermediate and thick cutaneous malignant melanoma. Preoperative and intraoperative lymph node mapping techniques were used to ascertain regional lymph node basins at risk for metastasis, and to identify the first node(s) the afferent lymphatics encounter in the basin, defined as the "sentinel" node(s). It has been shown that the histology of the sentinel node reflects the histology of the rest of the nodal bain, and according to preliminary studies using this technique, the likelihood of bypassing the sentinel node(s) to "higher" level nodes is less than 2%. Epidemiologic studies indicate that the long-term survival of patients with melanomas of intermediate thickness or greater is significantly compromised if regional lymph nodes are involved. Yet, the utility of performing lymph node dissections for the purposes of staging only is controversial, not only because of the morbidity and expense of the procedure, but the lack of proven survival benefit. In the present study, we performed preoperative and intraoperative lymphatic mapping, harvested clinically normal sentinel nodes, and examined them for micrometastasis by light microscopy. Both conventional stains and immunocytochemistry for S-100 protein and HMB-45 antibodies were performed, and only those patients with documented micrometastasis received complete lymph node dissections. The sentinel node(s) was identified in each of the patients. Micrometastatic disease was detected in 31 (23%) of the patients by selective lymphadenectomy, and the sentinel node(s) was the only node involved in 83% of the cases upon subsequent complete nodal dissection. Our preliminary results suggest that selective lymphadenectomy following lymphatic mapping is an effective procedure for staging melanoma patients with lesions of intermediate thickness or greater. Our results indicate that sentinel lymph nodes may be successfully identified and harvested in the majority of patients, and that they may be examined for the first evidence of micrometastasis without the need of a complete nodal dissection. Information as to whether micrometastases are present in the sentinel node would be valuable in staging patients, and identifying candidates for complete nodal dissections. We are participating in a National Cancer Institute-sponsored multicenter trial to ascertain whether this surgical approach can impact on the recurrence rate and survival of patients with stage 1 and 2 melanoma.

Identification of dendritic cell colony-forming units among normal human CD34+ bone marrow progenitors that are expanded by c-kit-ligand and yield pure dendritic cell colonies in the presence of granulocyte/macrophage colony-stimulating factor and tumor necrosis factor alpha.

Several cytokines, especially granulocyte/macrophage colony-stimulating factor (GM-CSF) and tumor necrosis factor alpha (TNF-alpha), have been identified that foster the development of dendritic cells from blood and bone marrow precursors in suspension cultures. These precursors are reported to be infrequent or to yield small numbers of dendritic cells in colony-forming assays. Here we readily identify dendritic cell colony-forming units (CFU-DC) that give rise to pure dendritic cell colonies. Human CD34+ bone marrow progenitors were expanded in semi-solid cultures with serum-replete medium containing c-kit-ligand, GM-CSF, and TNF-alpha. The addition of TNF-alpha to GM-CSF did not alter the number of typical GM colonies but did generate pure dendritic cell colonies that accounted for approximately 40% of the total colony growth. When the two distinct types of colonies were plucked from methylcellulose and tested for T cell-stimulatory activity in the mixed leukocyte reaction, the potency of colony-derived dendritic cells exceeded that of CFU-GM progeny from the same cultures by at least 1.5-2 logs. Immunophenotyping and cytochemical staining of the CFU-DC-derived progeny was also characteristic of dendritic cells. Other myeloid cells were not identified in these colonies. The addition of c-kit-ligand to GM-CSF- and TNF-alpha-supplemented suspensions of CD34+ bone marrow cells expanded CFU-DCs almost 100-fold by 14 d. We conclude that normal human CD34+ bone marrow cells include substantial numbers of clonogenic progenitors, distinct from CFU-GMs, that can give rise to pure dendritic cell colonies. These CFU-DCs can be expanded for several weeks by in vitro culture with c-kit-ligand, and their differentiation requires exogenous TNF-alpha in addition to GM-CSF. We speculate that this dendritic cell-committed pathway may in the steady state contribute cells to the epidermis and afferent lymph, where dendritic cells are the principal myeloid cell type, and may increase the numbers of these specialized antigen-presenting cells during T cell-mediated immune responses.