Sulforhodamine B interacts with albumin to lower surface tension and protect against ventilation injury of flooded alveoli

Abstract

In the acute respiratory distress syndrome, alveolar flooding by proteinaceous edema liquid impairs gas exchange. Mechanical ventilation is used as a supportive therapy. In regions of the edematous lung, alveolar flooding is heterogeneous, and stress is concentrated in aerated alveoli. Ventilation exacerbates stress concentrations and injuriously overexpands aerated alveoli. Injury degree is proportional to surface tension, T. Lowering T directly lessens injury. Furthermore, as heterogeneous flooding causes the stress concentrations, promoting equitable liquid distribution between alveoli should, indirectly, lessen injury. We present a new theoretical analysis suggesting that liquid is trapped in discrete alveoli by a pressure barrier that is proportional to T. Experimentally, we identify two rhodamine dyes, sulforhodamine B and rhodamine WT, as surface active in albumin solution and investigate whether the dyes lessen ventilation injury. In the isolated rat lung, we micropuncture a surface alveolus, instill albumin solution, and obtain an area with heterogeneous alveolar flooding. We demonstrate that rhodamine dye addition lowers T, reduces ventilation-induced injury, and facilitates liquid escape from flooded alveoli. In vitro we show that rhodamine dye is directly surface active in albumin solution. We identify sulforhodamine B as a potential new therapeutic agent for the treatment of the acute respiratory distress syndrome.

Keywords: acute respiratory distress syndrome; alveolar mechanics; rhodamine; surface tension; ventilation injury.

Fig. 1.

Pressures in a flooded alveolus. In the center of the alveolus, interface forms a meniscus of radius r_M and liquid-phase pressure is P_LIQ < alveolar air pressure P_ALV. At the border between 2 alveoli, interface has a saddle-shaped geometry. At this location, in the plane of the page, curvature is convex with interface border radius r_INT·BORD (inset); in a plane perpendicular to the page, curvature is concave with a larger radius approximately equal to duct radius r_DUCT (dashed gray arc). As liquid-phase pressure at the border between alveoli, P_LIQ·BORD, exceeds P_ALV (see text) and thus bulk liquid-phase pressure in flooded alveolus P_LIQ, there is a pressure barrier, ΔP_B, that traps liquid in the flooded alveolus. ΔP_B is proportional to surfact tension T.

Fig. 2.

Presence of sulforhodamine B (SRB) does not alter fluorescence intensity of fluorescein. Baseline images show isolated, perfused lung regions prepared for ventilation-injury assay. Perfusate is labeled with 23 μM fluorescein. Alveoli are flooded with 5% albumin in normal saline without or with 1 μM SRB. Inclusion of SRB does not alter alveolar liquid-phase fluorescence in green channel.

Fig. 3.

Rhodamine dye effects on surface tension in alveoli flooded with albumin solution. Flooding solution is albumin in normal saline plus 31 μM fluorescein and rhodamine dye—rhodamine WT (RWT), SRB, or sulforhodamine G (SRG)—as specified. Inflation increases surface tension: *P < 0.001 vs. transpulmonary pressure, P_ALV, of 5 cm H₂O data point in same group. Inclusion of RWT or SRB in flooding solution lowers surface tension at high lung inflation: #P < 0.01 vs. 4.6% albumin in normal saline at P_ALV of 15 cm H₂O.

Fig. 4.

SRB concentration effect on surface tension in alveoli flooded with albumin solution. Flooding solution is 4.6% albumin in normal saline plus 31 μM fluorescein and SRB as specified. Transpulmonary pressure is 15 cm H₂O. Inclusion of 1 nM–1 μM SRB in flooding solution lowers surface tension: *P < 0.01 vs. no SRB; #P < 0.02 vs. 1 nM and vs. 100 nM SRB.

Fig. 5.

Albumin concentration effect on surface tension in alveoli flooded with solution including lowest or highest effective SRB concentration. Flooding solution is 0–28% albumin in normal saline plus 31 μM fluorescein, without or with 1 nM (A) or 1 μM (B) SRB. Transpulmonary pressure is 15 cm H₂O. Albumin concentration of 12% alters surface tension: *P < 0.02 vs. normal saline without SRB. Albumin facilitates SRB surface activity: #P < 0.02 vs. same albumin concentration without SRB.

Fig. 6.

SRB surface activity in alveoli flooded with bovine or human albumin solution. Flooding solution is 4.6% bovine serum albumin (BSA) or human serum albumin (HSA) in normal saline plus 31 μM fluorescein and SRB as specified. Transpulmonary pressure is 15 cm H₂O. SRB lowers surface tension to the same degree in both albumin solutions: *P < 0.02 vs. same flooding solution without SRB.

Fig. 7.

SRB lessens ventilation injury of the alveolar-capillary barrier in regions with discrete alveolar flooding by albumin solution. Experiments are performed in isolated, perfused lung with 23 μM fluorescein included in the perfusate. Alveolar flooding solution is 3.0% albumin in normal saline, without or with 1 μM SRB. A: images of fluorescein fluorescence obtained at transpulmonary pressure of 5 cm H₂O at baseline (BL) before and at 11 min following 5 ventilation cycles with a positive end-expiratory pressure (PEEP) of 10 cm H₂O and a tidal volume, V_T, of 12 ml/kg. B: group injury score data for ventilation with PEEP and V_T as specified. Presence of discrete flooding causes injury: all discrete flooding groups differ (P < 0.001, statistics not shown on graph) from control, aerated groups. SRB can lessen injury: *P < 0.02 vs. same ventilation settings without SRB. Higher PEEP or V_T increases injury; among discrete flooding groups, a group with a letter at its base differs (P < 0.02) from all other groups except those with the same letter above their error bars.

Fig. 8.

RWT facilitates liquid escape from alveoli flooded with albumin solution. Flooding solution is 4% albumin in normal saline plus 25 μM 2′,7′-bis-(2-carboxyethyl)-5-(and-6)-carboxyfluorescein (BCECF) or 0.8 μM RWT. Ventilation is between P_ALV of 5 and 15 cm H₂O. Alveoli are imaged at P_ALV of 15 cm H₂O at ventilation cycle 0/BL and following cycles 20, 40, 60, 80, and 100. A: representative images of areas flooded with solution containing BCECF or RWT. Dashed lines enclose analysis areas comprising 31 ± 6 (n = 9, no difference between BCECF and RWT groups) alveoli, which are the same at all time points. F, alveolus that is flooded at all time points; A, alveolus that is aerated at all time points; *newly aerated alveolus that was flooded at previous time point. B: group data for % flooded alveoli in analysis region, normalized by % flooded at baseline. Baseline alveolar flooding averages 68 ± 11% (n = 9, no difference between groups). Inclusion of RWT promotes clearance. *P < 0.02 vs. baseline in same experimental group; #P < 0.02 vs. control, BCECF group at same time point.

Fig. 9.

Facilitation of SRB surface activity is specific to albumin. A: surface tension in isolated lung regions flooded with solutions containing specified solutes in normal saline, plus 31 μM fluorescein. Transpulmonary pressure is 15 cm H₂O. Only albumin facilitates SRB surface activity: *P < 0.01 vs. same solution in absence of SRB. B: injury score for isolated, perfused lung regions with heterogeneous alveolar flooding by solutions containing specified solutes in normal saline. Perfusate labeled with 23 μM fluorescein. Ventilation is with PEEP of 15 cm H₂O and V_T of 6 ml/kg. Albumin is necessary for SRB to lessen injury: *P < 0.02 vs. same solution without SRB.

Fig. 10.

SRB concentration effect on surface tension in alveoli flooded with blood plasma. Flooding solution is rat blood plasma plus 31 μM fluorescein and SRB as specified. Transpulmonary pressure is 15 cm H₂O. SRB lowers surface tension in plasma: *P < 0.02 vs. no SRB.

Fig. 11.

SRB is directly surface active in albumin solution. In vitro surface tension of 3-μl drops of normal saline with 31 μM fluorescein and solutes as specified. Albumin lowers surface tension: *P < 0.001 vs. normal saline solution without or with SRB. SRB further lowers surface tension: #P < 0.001 vs. albumin solution without SRB.

Fig. 12.

Variation in liquid-phase pressure and thickness at the edge of an alveolus. Figure shows free end of an interalveolar septum (black) surrounded by alveolar liquid (gray). An increase in pressure toward the end of the septum (pressure P closer to end greater than pressure p further from end) might be offset by a decrease in cross-sectional area of the liquid phase (area a closer to end less than area A further from end). If pA = Pa, then net force on the fluid element enclosed by dashed lines is 0.