1 of 50

Lecture 7: Density-dependent growth Part 1

Announcements:

  • HW1 due Monday 2/13/2023
  • HW2 posted, due 2/22/2023
  • Next Monday we will spend the lecture going through how to solve equations in python. If you’ve used sp.integrate() before and are familiar with it, you probably don’t need to come.

2 of 50

Lecture 7: Density-dependent growth Part 1

Today’s goals:

  • Review Monday’s lecture on exponential “growth laws”
  • Ask what happens to exponentially growing bacteria when they reach very high density and potentially exhaust nutrients
  • Develop a simple mathematical model for that scenario (see left graph)
    • The math associated with this model will be important for multiple, diverse subjects in the future

3 of 50

Quick review of last time

This network of metabolism and gene expression takes in nutrients

E. coli

and turns them into more biomass

The network is too complex to understand. But however it operates, it obeys these simple relationships!

 

 

Per-capita growth rate

Per-capita growth rate

Ribosomal content

4 of 50

Quick review of last time

Incorporating these measurements into a proteome allocation model allows you to precisely predict how the output of this network will change if you alter its inputs!

 

 

 

Simple relationship between nutrient conditions, metabolic proteins, and growth

Simple relationship between translational capacity, ribosome proteins, and growth

5 of 50

Today: back to our thought experiment

  • carbon (sugar, for example)
  • nitrogen (ammonia, glutamate, ...)
  • several salts (CaCl2, MgCl2, …)

6 of 50

Thought experiment

Liquid broth

?

7 of 50

+

8 of 50

Exponential growth

9 of 50

Exponential curves

10 of 50

Exponential growth not limited to microbes!

Collared Doves introduced to Great Britain after 1952

11 of 50

Also elephants . . .

Darwin, On the Origin of Species

Then why is the earth not overrun with elephants?

12 of 50

Why can’t exponential growth continue unabated?

M9 minimal medium

  • Glucose
  • MgSO4
  • CaCl2
  • Na2HPO4
  • KH2PO4
  • NaCl
  • NH4Cl
  • water

13 of 50

Why can’t exponential growth continue unabated?

It takes nutrients to build a cell.

Finite amount of nutrients, finite amount of cells!

Or high density of cells accumulates too much toxic byproducts

14 of 50

As always, let’s look at data!

Straight line on a log axis → what’s that?!

So what happens next?

OD remains constant for a while!

“Stationary phase”

The Bacterial Growth Curve!

15 of 50

A quick note:

For a system with a defined volume, like a bacterial culture tube, the population size and population density are directly proportional to each other!

 

 

 

Forgive me, I may use “population density” and “population size” independently!

16 of 50

Growth Phases:

  1. Lag phase: no growth
  2. Acceleration phase: growth rate increases
  3. Log phase or exponential phase: growth rate is constant
  4. Retardation phase: growth rate decreases
  5. Stationary phase: growth rate 0
  6. Decline phase: growth rate negative

17 of 50

Time

 

We have two parameters now:

  • Per capita growth rate/doubling time
  • Maximum cell density

We saw that the “quality” of the nutrient medium determines doubling time.

What determines the maximum cell density?

Let’s look at some experiments.

18 of 50

Experiment: what factors limit maximum cell density?

As we often do:

 

 

What do we see??

19 of 50

Cornelis Bernardus van Niel:�

  1. Grow “purple” bacteria under different initial concentrations of acetate
  2. Measure total growth

total biomass grown

Looks like total growth is directly proportional to initial carbon concentration!

 

total growth

initial nutrient concentration

proportionality constant that indicates how efficiently a nutrient is assimilated into biomass!

20 of 50

Total bacterial growth with acetate as a limiting nutrient (Athiorhodaceae)

(van Niel, Bact. Revs., 8, 1-118 (1944))

[acetate] (mg/ml)

0.5

1.0

2.0

3.0

total growth (mg/ml)

0.18

0.36

0.7

1.12

0.36

0.36

0.35

0.38

Total growth determined here by centrifuging final culture and measuring dry weight.

21 of 50

Efficiency specifically identified by isolating carbon

 

22 of 50

Total bacterial growth with acetate as a limiting nutrient (Athiorhodaceae)

(van Niel, Bact. Revs., 8, 1-118 (1944))

[acetate carbon] (mg/ml)

0.144

0.29

0.58

0.87

total carbon growth (mg/ml)

0.1

0.2

0.39

0.625

0.69

0.69

0.67

0.72

Total growth determined here by centrifuging final culture and measuring dry weight.

23 of 50

E. Coli grown with mannitol as the only source of carbon:

 

24 of 50

Time

We’re looking for a growth model that is exponential at early times/small population size

and then levels off.

We’ve seen how properties of the environment (nutrient quality and concentration) determine the properties of this growth curve.

Can we find a model for this growth dynamics?

25 of 50

“Density-dependent growth”

Time

 

We see that for this growth, the per capita growth rate changes as the culture density changes!

For this reason, this kind of growth is called “density-dependent growth”

 

26 of 50

How can we find a model for density-dependent growth?

For exponential growth, our key observation was:

One cell grows a little.

A lot of cells grow a lot.

Leading to a model where growth is proportional to population size!

 

Another way to visualize this model is by plotting the equation itself.

27 of 50

Population density vs. growth rate in exponential growth

How can we find such a relation for density-dependent growth?

As always, let’s first look at experimental data.

 

The differential equation is a formula that relates growth and cell density at any time!

28 of 50

Population Density

Let’s first compute growth rate, and then see how it depends on cell density to find a general formula relating growth and cell density!

29 of 50

derivative

Low cell density > little growth

High cell density > little growth

intermediate cell density > high growth

Population Density

Population Growth Rate

The right graph plots growth rate in time. Let’s now plot it as a function of cell density!

30 of 50

Low cell density: little growth

High cell density: little growth

intermediate cell density: high growth

Population Density

Population Growth Rate

↑ Our equation should look like this! ↑

31 of 50

Time

 

 

small N:

 

as N approaches some saturating value:

Something that is ~1 when N is small and 0 as N approaches some value

 

32 of 50

Time

 

 

 

 

Simplest:

 

 

 

 

 

 

Our model!! :

33 of 50

How does it look next to our data?

Experiment

Model

34 of 50

There are two phases for this kind of growth

 

 

 

35 of 50

 

 

 

 

36 of 50

 

There are 3 approaches to an equation:

  1. Engineer’s approach

“Just tell me the solution.”�

  • Physicist’s approach

“How do I solve it? Why do I believe it’s true?”�

  • Mathematician’s approach

“How much can I say about the solutions to the equation without ever solving it?”

Razvan Fetecau, my undergrad differential equations professor

Sounds funny, but this is a powerful approach!!

(We’ll solve it Monday in python!)

37 of 50

Back to our thought experiment

Let’s say we took an extremely dense bacterial culture, near saturation. What is its growth rate?

 

+

Now we add some cells to it.

What happens? Does its growth rate go way up?

 

 

38 of 50

Back to our thought experiment

+

Now we take media with no cells. What’s its growth rate?

 

What happens when you add a few cells to this?

 

!

 

39 of 50

One more experiment

We take our extremely dense culture again.

But now we dilute it with fresh media containing no cells.

What happens to the growth rate?

 

 

 

40 of 50

What is significant about these two situations in our model?

Tons of cells

No cells

 

 

 

 

 

 

“Steady states”

unstable

stable

41 of 50

How do you find steady states?

Look at the equation for your model!

 

 

 

 

42 of 50

Let’s do it!

 

 

 

 

 

 

43 of 50

µ

44 of 50

µ

45 of 50

µ

46 of 50

growth rate vs. N

 

 

 

etc..

µ

47 of 50

growth rate vs. N

 

 

 

etc..

 

 

 

 

µ

48 of 50

µ

49 of 50

 

What is the solution to this equation?

 

 

Let’s check the behavior we want to model:

Small t (small N):

 

 

t becomes large:

 

50 of 50