Does lymph have more proteins than plasma?

Pulmonary Edema from Increased Lung Protein Permeability in RDS

Animals that are born prematurely often die of respiratory failure from a condition that mimics the clinical, physiological and histological features observed in human infants with RDS (93, 145–147). Studies performed with chronically catheterized lambs that were delivered prematurely by caesarean section at ∼ 130 days gestation (term = 147 days) showed that respiratory failure developed in six of 10 lambs that were mechanically ventilated with 100% oxygen for 8 h after birth (93). These six lambs required peak inflation pressures that averaged > 50 cmH2O between 4 and 8 h after birth. They had severe hypoxemia and pulmonary hypertension, with a progressive increase in hematocrit and a reduction in plasma protein concentration secondary to generalized protein loss from the circulation. In contrast to earlier studies performed with more mature lambs (25), lung lymph flow and lymph protein flow remained high for the entire study. The postnatal tripling of lymph flow and lymph protein flow clearly showed that lung vascular permeability to protein increased in these preterm lambs with severe RDS (Fig. 7-7). Lung histology and postmortem measurement of extravascular lung water confirmed the presence of severe pulmonary edema (Fig. 7-8). In the four lambs that did not have RDS, lung lymph flow and protein clearance decreased to values that were at or below prenatal values, and postmortem measurements of extravascular lung water were significantly less than they were in lambs that had RDS. Thus, abnormal leakage of protein-rich liquid from the lung microcirculation into the interstitium constitutes a major component in the pathogenesis of RDS in preterm lambs (93). Subsequent studies showed that this lung vascular injury and edema can be inhibited by surfactant administration at birth (142), probably by reducing the need for high inflation pressures to achieve adequate ventilation and oxygenation, and by yielding uniform inflation of distal respiratory units (148).

It is noteworthy that mechanically ventilated preterm lambs with severe RDS had a marked reduction in circulating neutrophils within 30 min of birth, and that this was associated with abundant neutrophils in the lungs. The magnitude of the postnatal reduction in circulating neutrophils correlated with the degree of lung vascular protein leak and pulmonary edema (137). When lambs were rendered neutropenic from prenatal treatment with nitrogen mustard, lung vascular injury and edema did not develop postnatally after premature birth followed by 8 h of mechanical ventilation. These and earlier observations of neutrophil abundance in airway secretions of infants with severe RDS indicate that circulating neutrophils and their secretory products, specifically proteolytic enzymes and toxic oxygen metabolites, may play an important role in the pathogenesis of acute lung vascular protein leak and edema in RDS (137, 149–151). The mechanisms by which neutrophils are recruited into the lungs after premature birth and mechanical ventilation are unclear, but it is likely that several chemoattractants, including macrophage inflammatory protein-1α, and interleukins 6 and 8 (152–154), are released in response to the pulmonary stresses associated with increased blood flow and pressures within the lung circulation, and increased gas flow and pressures within the airways and distal air spaces.

2.2 The Contribution of the Parenchymal Proteome to Lymph Composition

Several proteomic analyses carried out on bovine, ovine, rodent, and human lymph sampled under physiological and pathological conditions have shown that the lymph fluid collects the “omic signature” of the parenchymal organ from which it drains (Clement et al., 2010, 2011, 2013; Clement and Santambrogio, 2013; D'Alessandro et al., 2014; Dzieciatkowska et al., 2011, 2014; Veenstra, 2007; Veenstra et al., 2005; Zurawel et al., 2011). Although plasma albumin and serum globulins constitute the majority of the lymph proteins, tissue-specific proteins are also highly represented in the lymph proteome when compared to the plasma proteome. The most striking differences between lymph and plasma proteomes are: (i) extracellular matrix (ECM) proteins and products from their processing, derived from tissue growth and remodeling, are more highly represented in the lymph compared to the plasma, as well as, (ii) proteins resulting from ongoing cellular metabolic/catabolic activities in each parenchymal organ, and (iii) intracellular proteins released from apoptotic cells (D'Alessandro et al., 2014; Fang et al., 2010; Goldfinch et al., 2008; Meng and Veenstra, 2007; Mittal et al., 2009; Nguyen et al., 2010; Zurawel et al., 2011).

On the other hand, proteins pivotal to the maintenance of the intracapillary osmotic pressure (albumin, and α1, α2, β globulins) and coagulation factors are more highly represented in the plasma than the lymph proteome.

Unburned Tissue

Generalized edema in soft tissues not directly injured is another characteristic of large cutaneous burns. Brouhard et al.53 reported increased water content in unburned skin even after a 10% total body surface area (TBSA) burn, with the peak edema occurring 12 hours postburn. Arturson reported an increased transvascular fluid flux (lymph flow) from unburned tissue and a transient increase in permeability as measured by an increase in the lymph concentration of plasma protein and macromolecular dextran infused as a tracer.14,34 Harms et al.35 extended these findings by measuring changes in lymph flow and protein transport in noninjured soft tissue for 3 days after injury. They found that skin and muscle permeability (flank lymph from sheep) were elevated for up to 12 hours postburn for molecules the size of albumin and immunoglobulin G, but the microvascular permeability of the lung (lymph from caudal mediastinal node) showed no increase. Maximum increased lymph flow and tissue water content were observed to correlate with the severe hypoproteinemia that occurred during the early resuscitation period of a 40% burn injury in sheep.7,54

The mechanism by which burn injury induces microvascular hyperpermeability remote from the injury has been the subject of extensive study utilizing plasma from burned animals transfused into unburned recipients by Kremer et al.55 These transfusions cause endothelial activation, albumin leakage, and leukocyte adhesion and rolling via an undefined transfused circulating factor. This model allows testing of the effects of various inhibitory and therapeutic agents in ameliorating the resultant physiologic derangements. High-dose vitamin C, an antioxidant, administered to the recipient rat was found to significantly reduce capillary leakage but not leukocyte-endothelial interactions.55,56 Ketanserin, a 5-HT2a antagonist, reduced plasma extravasation and leukocyte-endothelial interactions after burn plasma transfer.57 These findings were further confirmed with the serotoninergic receptor-blocking agents cinanserin and methysergide.58,59 When the burn wounds of donor burned rats were bathed in cerium nitrate, a topical antimicrobial and putative anti-inflammatory agent, the resulting plasma no longer induced injury in transfused rats.60 In further burn plasma transfer studies Hernekamp et al.61,62 also demonstrated that the cholinergic anti-inflammatory pathway stimulated by cbp-choline or pretreatment with physostigmine can similarly attenuate albumin efflux and leukocyte adhesions.

Changes in endothelial permeability

Burn injury, trauma and sepsis increase microvascular permeability in both the systemic101,102 and the pulmonary circulation.103 Increased vascular permeability and cutaneous edema are hallmarks of burn shock. The increase in systemic vascular permeability results in the exudation of protein-rich fluid from the vascular compartment into the interstitium. This results in intravascular volume loss and the concomitant development of interstitial edema. If resuscitation is not prompt and adequate, this loss in intravascular volume will lead to intravascular hypovolemia, hypotension, and inadequate tissue perfusion. Severe interstitial edema formation can lead to the development of compartment syndrome, with compromise of neurovascular integrity. The lungs may also be affected. Edema formation due to an increase in microvascular permeability is a hallmark of the acute lung injury. The factors that determine the transvascular fluid flux are summarized in the Starling–Landis equation:104,105

Jv=Kf[(Pmv−Pi)−α(πmv−πI)]

where Jv is the transvascular fluid flux, Kf is the filtration coefficient (measure of the endothelial permeability to small solutes and water as well as of the permeability surface area), Pmv is the microvascular hydrostatic pressure, Pi is the interstitial hydrostatic pressure, α is the osmotic reflection coefficient to protein, πmv is the microvascular oncotic pressure, and πi is the interstitial oncotic pressure.

Several investigators have studied lymph flow and lymph protein flux after administration of bacteria or short-time infusion of 1–2 µg/kg of endotoxin in sheep. Two phases of permeability change could be distinguished in these models.106 During phase 1 there was a high microvascular hydrostatic pressure as defined by the Gaar equation.107 This was associated with an increase in lung lymph flow, but the lymph protein concentration was low. It was concluded that the high microvascular hydrostatic pressure was responsible for the early increase in transvascular fluid flux. Thromboxane A2 (TXA2) has been found to be responsible for the vasoconstriction that causes local increases in hydrostatic pressure. Therefore, it is not surprising that administration of the thromboxane synthetase inhibitor OKY046 prevented the rise in lymph flow during phase 1.108 That effect was also noted after blockage of cyclooxygenase by ibuprofen.109 Early edema formation at the site of burn injury might be due to a different mechanism. Recent data suggest that a marked fall in interstitial hydrostatic pressure might occur in the injured tissue, which could explain the immediate onset of edema formation after thermal injury.110,111 These changes might be the result of an inhibition of the fibroblast β1-integrin attachment to collagen.

During phase 2, lymph flow continues to be high. However, the lymph protein concentration rises considerably and the pulmonary artery pressure is only mildly elevated.106 The oncotic pressure gradient between microvasculature and interstitial space is reduced during that period.112 Together, these data suggest that the permeability of the pulmonary endothelium to protein increases in phase 2. In fact, the reflection coefficient for total protein fell from 0.73 to 0.58, with respective changes in the reflection coefficients for albumin (0.66 to 0.5), IgG (0.76 to 0.64), and IgM (0.91 to 0.83) after 4 hours of Escherichia coli sepsis in sheep.113 Confirmation of this hypothesis is still pending in models of endotoxemia, but it has been generally accepted that the changes in pulmonary transvascular fluid flux in phase 2 represent changes in microvascular permeability. The mechanisms of the increased microvascular permeability are still under discussion.

Endothelial cells play an important role in the regulation of vascular permeability. It has been hypothesized that endothelial cells can contract upon stimulation.114 As a result, the intercellular gaps might increase in number and/or size, establishing the so-called capillary leak syndrome. The development of the protein-rich high-permeability edema can be ameliorated if substances are administered that raise the endothelial cell content of cyclic adenosine or guanosine monophosphate.115,116 However, endothelial cells do not merely serve as targets during systemic inflammation: they actively contribute to the ongoing inflammatory process. The endothelial cell can be stimulated by endotoxin, TNFα, or IL-1 to express E-selectin, an adhesion molecule.117 E-selectin on the surface of endothelial cells interacts with the corresponding l-selectin complex on PMNs to facilitate rolling of these cells along the endothelium.118 Moreover, endothelial cells secrete the proinflammatory cytokines TNFα and IL-1, which activates PMNs.119 Conflicting data exist regarding the role of PMNs in SIRS. PMNs are usually found at the site of tissue injury, to which they migrate following a concentration gradient of chemotactic stimuli. Upon stimulation, PMNs roll along endothelial cells, and in a further step mediated by PMN CD18/CD11b interactions with endothelial ICAM-1, emigrate from the vessel into the interstitial space. Antibodies against the common CD18β chain showed beneficial effects in an animal model of sepsis-induced lung injury, suggesting that integrin-mediated PMN emigration is a functionally important process in the development of acute lung injury.120 On the other hand, patients who are deficient in CD18 have abundant PMNs in their alveolar spaces, and the monoclonal antibody 60.3 was ineffective in completely blocking the migration of PMNs into the lung in a number of conditions.121 We have reported that in chronic endotoxemia there were few PMNs in the lung but numerous macrophages.122 Activated PMNs and macrophages release oxygen free radicals and proteases at sites of inflammation. Those processes appear to be functionally important in the development of vascular permeability because administration of oxygen free radical scavengers and antiproteases proved to be useful in diminishing edema accumulation after endotoxin challenge.123,124 However, proteases and oxygen radicals are also released by macrophages, which are already present in the tissue, and by monocytes that migrate to sites of inflammation. Depletion of granulocytes by anti-PMN antiserum or by treatment with nitrogen mustard did not prevent the changes in microvascular permeability following the administration of endotoxin.125,126 Moreover, patients deficient in PMNs would still develop the adult respiratory distress syndrome (ARDS) associated with sepsis.126,127 On the other hand, treatment of sheep or goats with hydroxyurea, which is another compound used to deplete granulocytes, was effective and diminished fluid accumulation in the lung after endotoxin challenge.128 However, urea scavenges free radicals, which might explain its efficacy.129 As the inflammatory response becomes chronic, many mediators have been released and more than one mechanism might be assumed to be responsible for the capillary leak.

The role of arachidonic acid metabolites in facilitating increased vascular permeability has been extensively investigated. Administration of the thromboxane synthetase inhibitor OKY046 not only reduced transvascular fluid flux in phase 1, but was also effective in phase 2 after endotoxin challenge.108 This finding suggests that thromboxanes participate in permeability alterations during systemic inflammation. Oxygen free radicals can also increase microvascular permeability, both by activation of endothelial cell contraction and by damaging the endothelial cell membrane.130 OKY046 has been shown to reverse oxygen free radical-induced lung injury.131,132 On the other hand, inhibition of the cyclooxygenase did not affect transvascular fluid flux during phase 2, even though thromboxane A2 is a cyclooxygenase metabolite.109 This discrepancy is still unexplained; however, prostacyclin is elevated after endotoxin administration, and this material has many actions that counter the actions of thromboxane. Administration of a cyclooxygenase inhibitor will prevent the release of this salutary eicosanoid.

TNFα is one of the early mediators in systemic inflammation. It has been reported to be elevated during sepsis and endotoxemia after hemorrhagic shock or thermal injury. It is considered to be one of the most important mediators in the cascade because it has the potential to stimulate or enhance most of the steps in the inflammatory response. Moreover, administration of human recombinant TNFα reproduced most of the effects of endotoxemia, including alterations in pulmonary microvascular permeability in the chronic sheep model.133,134 TNFα also induces the secretion of PAF, which is a further early mediator of systemic inflammation. PAF causes an increase in lung lymph flow and permeability to protein when it is infused into conscious sheep.135 Administration of a PAF antagonist abolished the cardiopulmonary response that occurs during phase 1 and attenuated it during phase 2.136 However, PAF had no direct effect on endothelial cells. This suggests that it probably increases microvascular permeability through other mechanisms, such as its priming effect on PMNs.137,138

If burn patients are adequately resuscitated, they most commonly enter into a phase characterized by hyperdynamic cardiovascular function. The hyperdynamic cardiovascular response is associated with profound changes in pulmonary transvascular fluid flux in the ovine model of continuous endotoxemia.102,139 The lymph protein concentration gradually decreased after phase 2, and after 24 hours of endotoxemia the reflection coefficient to protein was at baseline level, whereas the lymph flow was still high. Microvascular hydrostatic pressure, evaluated by Holloway's technique, was not significantly different from baseline.140 The elevated transvascular fluid flux was attributed to a high filtration coefficient. An increase in both perfused surface area and pore numbers might have contributed to the change in filtration. Repeated injections of endotoxin also reduced subsequent lung lymph production in response to endotoxin.133 These changes in lung lymph flow were associated with elevations in endothelin and atrial natriuretic peptide.140,141 However, further studies must determine whether these factors affect pulmonary microvascular changes during the late phases of sepsis and multiple organ failure.

Non-burned tissue

Generalized edema in soft tissues not directly injured is another characteristic of large cutaneous burns. Brouhard et al.65 reported increased water content in non-burned skin even after a 10% burn, with the peak edema occurring 12 hours post burn. Arturson reported an increased transcapillary fluid flux (lymph flow) from non-burned tissue and a transient increase in permeability, as measured by an increase in the lymph concentration of plasma protein and macromolecular dextran infused as a tracer.15,21,59 Harms et al.56 extended these findings by measuring changes in lymph flow and protein transport in non-injured soft tissue for 3 days after injury. They found that skin and muscle permeability (flank lymph from sheep) were elevated for up to 12 hours post burn for molecules the size of albumin and immunoglobulin G, but the microvascular permeability of the lung (lymph for caudal mediastinal node) showed no increase. Maximum increased lymph flow and tissue water content were observed to correlate with the severe hypoproteinemia that occurred during the early resuscitation period of a 40% burn injury in sheep.8,61 The sustained increase in water content and the elevated lymph flow of the non-burned tissue after the return of normal permeability is likely the result of the sustained hypoproteinemia.51,56,59,60

Demling and colleagues66 suggested that the edema could be partially attributed to alterations in the interstitial structure. They suggested that interstitial protein washout increases the compliance of the interstitial space and that water transport and hydraulic conductivity across the entire blood–tissue–lymph barrier increased with hypoproteinemia. Several clinical and animal studies have established that maintaining higher levels of total plasma protein concentration can ameliorate the overall net fluid retention and edema.7,67 Non-burn edema can also be moderated by infusion of non-protein colloids such as dextran, if the colloid osmotic gradient is increased above normal.8,61 However, it is not known whether either the correction of hypoproteinemia or the use of either albumin or dextran leads to improved clinical outcome. It has been reported that the use of colloids has no beneficial effect on edema in the burn wound.50,61 Use of hypertonic saline formulations as initial fluid therapies for burn shock can greatly reduce initial volume requirements and net fluid volume (infused in minus urine out).68,69 However, a rebound of fluid requirements and net fluid can occur after early use of hypertonics and colloids.61,68 Retrospective analyses of patients correlating early albumin use with fluid requirements show significant volume sparing during the first post-burn day, but after 48 hours the effect is less apparent.70

INHALATION INJURY

Inhalation injury is a major cause of death in burn patients. The mortality rate for children with isolated cutaneous burns is 1% to 2% but increases to approximately 40% in the presence of inhalation injury.49,92 Inhalation injury is caused primarily by inhaled toxins such as fumes, gases, and mists. Although the supraglottic region can be injured by both thermal and chemical insults, tracheobronchial and lung parenchymal injuries rarely occur as a result of direct thermal damage because the heat disperses rapidly in the larynx. Hypoxia, increased airway resistance, decreased pulmonary compliance, increased alveolar epithelial permeability, and increased pulmonary vascular resistance may be triggered by the release of vasoactive substances (thromboxane A2, C3a, and C5a) from damaged epithelium.93 Neutrophil activation plays a critical role in this process, and pulmonary function has been shown to improve with the use of a ligand binding to E-selectins (inhibiting neutrophil adhesion) and anti-interleukin-8 (inhibiting neutrophil chemotaxis). Another significant form of respiratory tract pathology is the sloughing of ciliated epithelial cells from the basement membrane, resulting in exudate formation. The exudate, which consists of lymph proteins, coalesces to form fibrin casts. These fibrin casts are frequently resistant to routine pulmonary toilet and can create a “ball-valve” effect in localized areas of lung, eventually causing barotrauma.

The diagnosis is usually made based on clinical history and physical examination findings during the initial evaluation. Victims trapped in a house fire with excessive smoke and fumes are likely to have sustained severe inhalation injury. Facial burns with singed hair and carbonaceous sputum suggest the presence of inhalation injury. Hoarseness and stridor suggest significant airway obstruction, so the airway should be secured immediately with endotracheal intubation. Patients who present with disorientation and obtundation are likely to have elevated carbon monoxide levels (carboxyhemoglobin >10%). Cyanide toxicity as a result of the combustion of common household items may also contribute to unexplained metabolic collapse. Diagnostic tools, such as bronchoscopy and xenon-133 scanning, are more than 90% accurate in determining the presence of inhalation injury. Fiber-optic bronchoscopic examination of the airway at the bedside (avoiding the need to transport critically injured burn patients to the nuclear medicine department) is usually sufficient to identify airway edema and inflammatory changes of the tracheal mucosa such as hyperemia, mucosal ulceration, and sloughing. It remains the gold standard to confirm the presence of inhalation injury.56 Ventilation scan with xenon 133 can also identify regions of inhalation injury by assessing respiratory exchange and excretion of xenon by the lungs.76

The treatment of inhalation injury begins at the scene of the burn accident. The administration of 100% oxygen rapidly decreases the half-life of carbon monoxide. The airway must be secured with intubation in patients exhibiting signs and symptoms of imminent respiratory failure. Aggressive pulmonary toilet with physiotherapy and frequent suctioning is vital to prevent any serious respiratory complications. Humidified air is delivered at high flow, and bronchodilators and racemic epinephrine are used to treat bronchospasm. IV heparin has been shown to reduce tracheobronchial cast formation, improve minute ventilation, and lower peak inspiratory pressures after smoke inhalation. Inhalation treatments such as 20% acetylcysteine nebulized solution (3 mL every 4 hours) plus nebulized heparin (5000 to 10,000 units with 3 mL normal saline every 4 hours) are effective in improving the clearance of tracheobronchial secretions and minimizing bronchospasm, thereby significantly decreasing reintubation rates and mortality.10,62

The presence of inhalation injury generally requires increased fluid resuscitation, up to 2 mL/kg per percent TBSA burned more than would be required for the same size burn without an inhalation injury. In fact, pulmonary edema that is associated with inhalation injury is not prevented by fluid restriction; rather, inadequate resuscitation may increase the severity of pulmonary injury by the sequestration of polymorphonuclear cells.40 Steroids have not been shown to be of any benefit in inhalation injury. Prophylactic IV antibiotics are not indicated but are started if there is a clinical suspicion of pneumonia. Early pneumonia is usually the result of gram-positive organisms such as methicillin-resistant S. aureus, whereas later infection is caused by gram-negative organisms such as Pseudomonas. Serially monitored sputum cultures and bronchial washings should guide antibiotic therapy.

Starling Pressures and Local Lymph Flow

In most tissues, ISF volume is maintained by a low level of filtration from the microcirculation being matched by an equal efflux of fluid from the tissue in the lymph. From this, one would anticipate that the magnitude of the lymph flow from a tissue could be estimated from the mean Starling pressures, the LP, and the exchange surface area of the microcirculation. In an incisive review, Levick62 pointed out that where the mean Starling pressures have been measured, they predict much greater filtration rates than the lymph flow and the lymph protein concentration would indicate. From 15 sets of data, he estimated P0, the net pressure opposing filtration from the following relation:

(9.19)P0=σm(Πc−Πi)+Pi

When values of P0 are compared with direct measurements of pressure in post-capillary venules, PVc, it is found that PVc is nearly always greater than P0, and in several cases PVc exceeds P0 by 5 to 10 mmHg (Table 9.2). Net driving pressures for filtration of this magnitude would be expected to result in high lymph flows, but in most of the tissues concerned (e.g., subcutaneous tissue, muscle, and mesentery), the basal lymph flows are so low that they are difficult to measure.

Table 9.2. Starling Pressures in Muscle, Mesentery, and Subcutaneous Tissues at Heart Level

Species and TissueΠcΠiPiPoPcPc−PoDog, skeletal muscle26.011.0−2.013.012–20−1+7Cat, mesentery19.16.1013.023.510.5Human, chest subcutis26.815.6−1.59.7>15.0>5.3

Po is calculated from Eq. (9.17) assuming that σ = 1 and thus is an overestimate. Pc is based on direct measurements in venules or venular capillaries, and is therefore an underestimate of mean Pc. The difference (Pc−Po) is consequently the minimum difference based on available data.

From Levick, J. R. (1991). Capillary filtration–absorption balance reconsidered in light of dynamic extravascular factors. Exp. Physiol. 76, 825–85.60

To account for these discrepancies, Levick62 suggested that vasomotion, the spontaneous contraction and relaxation of arteriolar smooth muscle, might be responsible for large variations in Pc, and that Pc measurements tended to be made in vessels where there was brisk flow and a higher than average Pc. The few direct estimates of Pc during vasomotion, however, suggest that the fluctuations are relatively small.

An alternative hypothesis was put forward by Michel79 and Weinbaum.132 Independently, they both realized that if the filtrate leaving the “small pores” was uncontaminated by fluid containing the higher concentration of macromolecules, the effective σΔΠ opposing filtration may approximate to that across the small pores. This would be considerably greater than that calculated from global values of Πi, and would increase the force opposing filtration (P0), reducing filtration rates to levels consistent with basal rates of lymph flow. The significant deviations between P0 and PVc reported by Levick62 are found in microvessels with continuous endothelium, where the small pores are the interstices of the glycocalyx lying above the intercellular clefts. The downstream side of the microvascular ultrafilter is therefore the abluminal surface of the glycocalyx, which is separated from the ISF immediately outside the microvessel by the intercellular cleft with its tortuous pathway through the tight junctions. The pathway for macromolecules (the large pores) is either through the endothelial cell vesicles (via channels or by a fission–fusion mechanism) or by the very occasional leaky intercellular cleft. For the high protein concentration of this large pore filtrate to mix with that emerging from the small pores, protein molecules would have to diffuse back through the intercellular clefts to the site of ultrafiltration at the glycocalyx. Although this pathway is short, diffusion has to occur against the flow of fluid from the vessel lumen. The velocity of this filtrate is increased ten-fold or more as it passes through the breaks in the tight junctions. Rough calculations79 and a detailed mathematical model55 both indicate that, even with filtration rates driven by pressure differences across the glycocalyx as small as 1–2 cm H2O, the fluid velocity through the breaks in the tight junctions impose a major barrier to the diffusion of proteins through the clefts in the luminal direction. These levels of filtration, nevertheless, are consistent with basal rates of lymph flow.

The hypothesis has been examined experimentally in single frog mesenteric capillaries54 and in rat mesenteric venules.3 In these studies, it was found that even when the interstitial concentration of serum albumin in contact with the outside wall of a vessel was the same as that in the perfusate, fluid movements through the vessel wall were opposed by oncotic pressures much greater than those estimated from global values of Πi. The authors concluded that these observations were consistent with the oncotic pressures opposing fluid filtration from continuous (non-fenestrated) capillaries are developed across the glycocalyx, and that global values of Πi do not determine fluid exchange directly. Some of their data are shown in Figure 9.18.

Transient and steady-state fluid exchange and their relations to local lymph flow have been reviewed recently by Levick and Michel.64

Infectious Diseases

Multiple lines of B cell and T-cell-deficient mice have been used for the study of B cell (humoral) and T-cell (cellular) responses for many diseases. Compared to humans, pigs are often infected by closely related viral and bacterial pathogens. They can serve as zoonotic reservoirs for human pathogens. For major enteropathogens, mouse models do not reproduce the tropism and immunopathology of the corresponding human infections. Thus, experiments in pigs may contribute to the study of the basic immune response to a pathogenic challenge as well as to discriminate cellular and humoral protective immunity to infectious agents.

Rodents and humans can utilize four to six functional joining gene segments (JH) of the heavy chain (HC) locus, whereas pigs only possess one functional JH locus. Targeting the single functional joining region (JH) of the HC locus in porcine fibroblasts of the Large White breed was carried out by homologous recombination with a knockout vector removing the single functional JH sequences. SCNT using one targeted cell clone produced heterozygous knockout animals that were mated to produce homozygous knockout piglets. The homozygous knockouts appeared in the litters with the expected Mendelian ratio. mRNA analysis in spleen and mesenteric lymph nodes as well as protein analysis by flow cytometry detected that homozygous knockout piglets did not show mature B cells and antibodies of any isotype. No circulating IgM+ B cells were observed in the homozygous knockouts, showing that the VDJ rearrangement is impaired. Functional VDJ rearrangement at the HC locus has been shown to be required for B cell development. No transcription or secretion of immunoglobulin isotypes was observed in blood and secondary lymphoid tissues. No lymph node follicles were found. The findings were consistent with the results in B cell-deficient mice and humans. Heterozygous knockouts showed IgM+ B cells of more than half of the amount of wild-type controls. In contrast, T-cell receptor (Tcr)β transcription and T-cells seemed to be largely unaffected in homozygous knockouts.

The homozygous knockout pigs were conventionally reared by allowing them to suckle until weaning 4 weeks after birth. In the suckling period, they remained healthy; however, thereafter, they were severely susceptible to bacterial infections leading to a wasting-like syndrome and the impossibility to conventionally maintaining them longer than a few weeks after weaning. Thus, gnotobiotic facilities have to be used for maintaining the homozygous knockouts. These antibody- and B cell-deficient pigs may be used as the first step in the generation of HC knockout pigs for the subsequent complete replacement of the porcine immunoglobulin loci with the human homologous genomic sites. These genetically engineered animals in the future might also serve as a source of pathogen specific human polyclonal antibodies thereby alleviating both the supply and specificity issues as polyclonal antibodies harvested from human donors by plasmapheresis are of broad specificity and the supply is limited.140

Therefore, the next step for this aim was carried out. The porcine κ light chain immunoglobulin locus is in the megabase range and comprises multiple families of variable (Vκ) genes, five joining (Jκ) regions, and—as found for most common mammals including the mouse—a single constant (Cκ) gene. To inactivate the pig κ locus, the single Cκ gene offered the best option, as this strategy has been proven to be successful in the mouse. Therefore, targeted deletion of the most part of the porcine κ light chain constant (Cκ) region was carried out in pig primary fetal fibroblasts. By SCNT, heterozygous knockouts were produced with one mutant cell clone. Heterozygous knockouts were bred together to generate homozygous knockouts at a ratio close to that expected by the Mendelian inheritance. Peripheral blood mononuclear cells as circulating B cells and secondary lymphoid organs like mesenteric lymph nodes—which contain follicles that consist of a dense population of B cells—from homozygous knockouts were devoid of κ-containing immunoglobulins. Furthermore, there was an increase in λ light chain expression when compared to that of wild-type littermates. Tcrβ rearranged gene transcription was not altered.

Why lymph has less protein than plasma?

What contains more plasma or lymph?

Does the lymph contain proteins?

How lymph is different from plasma?